id
stringlengths
9
10
submitter
stringlengths
1
64
authors
stringlengths
4
7.24k
title
stringlengths
1
278
comments
stringlengths
1
762
journal-ref
stringlengths
1
557
doi
stringlengths
11
153
report-no
stringlengths
2
254
categories
stringlengths
5
98
license
stringclasses
9 values
abstract
stringlengths
6
3.8k
versions
list
update_date
timestamp[s]
authors_parsed
list
predictions
stringclasses
2 values
probabilities
float64
0.5
1
1004.3702
Lizhi Du
Lizhi Du
A Polynomial time Algorithm for Hamilton Cycle with maximum Degree 3, 3SAT
16 pages. This time, I add a detailed polynomial time algorithm and proof for 3SAT
null
null
null
cs.DS
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Based on the famous Rotation-Extension technique, by creating the new concepts and methods: broad cycle, main segment, useful cut and insert, destroying edges for a main segment, main goal Hamilton cycle, depth-first search tree, we develop a polynomial time algorithm for a famous NPC: the Hamilton cycle problem. Thus we proved that NP=P. The key points of this paper are: 1) there are two ways to get a Hamilton cycle in exponential time: a full permutation of n vertices; or, chose n edges from all k edges, and check all possible combinations. The main problem is: how to avoid checking all combinations of n edges from all edges. My algorithm can avoid this. Lemma 1 and lemma 2 are very important. They are the foundation that we always can get a good branch in the depth-first search tree and can get a series of destroying edges (all are bad edges) for this good branch in polynomial time. The extraordinary insights are: destroying edges, a tree contains each main segment at most one time at the same time, and dynamic combinations. The difficult part is to understand how to construct a main segment's series of destroying edges by dynamic combinations. The proof logic is: if there is at least on Hamilton cycle in the graph, we always can do useful cut and inserts until a Hamilton cycle is got. The times of useful cut and inserts are polynomial. So if at any step we cannot have a useful cut and insert, this means that there are no Hamilton cycles in the graph. In this version, I add a detailed polynomial time algorithm and proof for 3SAT
[ { "version": "v1", "created": "Mon, 12 Apr 2010 04:39:27 GMT" }, { "version": "v10", "created": "Mon, 5 Nov 2012 01:44:46 GMT" }, { "version": "v11", "created": "Thu, 31 Jan 2013 11:15:53 GMT" }, { "version": "v12", "created": "Mon, 4 Nov 2013 14:09:42 GMT" }, { "version": "v13", "created": "Thu, 7 Aug 2014 02:01:59 GMT" }, { "version": "v14", "created": "Wed, 20 Aug 2014 01:20:04 GMT" }, { "version": "v15", "created": "Tue, 2 Sep 2014 15:17:04 GMT" }, { "version": "v16", "created": "Thu, 18 Sep 2014 12:10:37 GMT" }, { "version": "v17", "created": "Mon, 13 Oct 2014 03:52:56 GMT" }, { "version": "v18", "created": "Mon, 17 Nov 2014 10:54:13 GMT" }, { "version": "v19", "created": "Mon, 1 Dec 2014 11:05:59 GMT" }, { "version": "v2", "created": "Tue, 23 Nov 2010 22:00:44 GMT" }, { "version": "v20", "created": "Mon, 22 Dec 2014 04:48:29 GMT" }, { "version": "v21", "created": "Mon, 11 May 2015 12:11:22 GMT" }, { "version": "v22", "created": "Thu, 24 Sep 2015 13:53:17 GMT" }, { "version": "v23", "created": "Mon, 30 Nov 2015 04:27:21 GMT" }, { "version": "v24", "created": "Mon, 21 Mar 2016 14:41:54 GMT" }, { "version": "v25", "created": "Mon, 13 Jun 2016 06:25:23 GMT" }, { "version": "v26", "created": "Fri, 5 Aug 2016 18:15:23 GMT" }, { "version": "v27", "created": "Mon, 29 Aug 2016 02:17:42 GMT" }, { "version": "v28", "created": "Thu, 10 Nov 2016 05:51:30 GMT" }, { "version": "v29", "created": "Mon, 23 Jan 2017 14:35:27 GMT" }, { "version": "v3", "created": "Mon, 10 Jan 2011 02:50:16 GMT" }, { "version": "v30", "created": "Tue, 28 Feb 2017 13:16:20 GMT" }, { "version": "v31", "created": "Wed, 22 Mar 2017 12:20:11 GMT" }, { "version": "v32", "created": "Tue, 11 Apr 2017 08:21:53 GMT" }, { "version": "v33", "created": "Mon, 19 Jun 2017 11:57:26 GMT" }, { "version": "v34", "created": "Wed, 17 Jan 2018 12:32:05 GMT" }, { "version": "v35", "created": "Tue, 13 Feb 2018 04:04:57 GMT" }, { "version": "v36", "created": "Mon, 12 Mar 2018 11:17:48 GMT" }, { "version": "v37", "created": "Mon, 11 Jun 2018 01:14:42 GMT" }, { "version": "v38", "created": "Wed, 11 Jul 2018 10:27:51 GMT" }, { "version": "v39", "created": "Mon, 30 Jul 2018 22:02:52 GMT" }, { "version": "v4", "created": "Sun, 1 May 2011 01:32:02 GMT" }, { "version": "v40", "created": "Tue, 21 Aug 2018 00:01:46 GMT" }, { "version": "v41", "created": "Sun, 2 Sep 2018 23:24:26 GMT" }, { "version": "v42", "created": "Tue, 18 Sep 2018 07:54:59 GMT" }, { "version": "v43", "created": "Wed, 24 Oct 2018 01:58:41 GMT" }, { "version": "v44", "created": "Thu, 7 Feb 2019 04:25:15 GMT" }, { "version": "v45", "created": "Thu, 21 Mar 2019 11:10:18 GMT" }, { "version": "v46", "created": "Thu, 2 May 2019 01:32:57 GMT" }, { "version": "v47", "created": "Mon, 24 Jun 2019 00:56:03 GMT" }, { "version": "v48", "created": "Thu, 10 Oct 2019 07:09:08 GMT" }, { "version": "v49", "created": "Sun, 17 Nov 2019 01:38:36 GMT" }, { "version": "v5", "created": "Fri, 7 Oct 2011 01:39:26 GMT" }, { "version": "v50", "created": "Thu, 23 Jan 2020 05:49:11 GMT" }, { "version": "v51", "created": "Mon, 27 Apr 2020 00:24:06 GMT" }, { "version": "v52", "created": "Sun, 7 Jun 2020 21:56:02 GMT" }, { "version": "v53", "created": "Mon, 6 Jul 2020 10:07:25 GMT" }, { "version": "v54", "created": "Sun, 2 Aug 2020 22:55:43 GMT" }, { "version": "v55", "created": "Wed, 2 Sep 2020 01:02:09 GMT" }, { "version": "v56", "created": "Thu, 8 Oct 2020 01:05:54 GMT" }, { "version": "v57", "created": "Tue, 10 Nov 2020 14:01:31 GMT" }, { "version": "v58", "created": "Thu, 3 Dec 2020 06:27:25 GMT" }, { "version": "v59", "created": "Wed, 20 Jan 2021 11:52:23 GMT" }, { "version": "v6", "created": "Fri, 6 Apr 2012 11:16:37 GMT" }, { "version": "v60", "created": "Tue, 2 Feb 2021 01:58:47 GMT" }, { "version": "v61", "created": "Thu, 8 Apr 2021 07:36:54 GMT" }, { "version": "v62", "created": "Mon, 10 May 2021 00:01:29 GMT" }, { "version": "v63", "created": "Tue, 3 Aug 2021 12:02:09 GMT" }, { "version": "v64", "created": "Thu, 30 Sep 2021 08:07:36 GMT" }, { "version": "v65", "created": "Thu, 4 Nov 2021 13:33:17 GMT" }, { "version": "v66", "created": "Tue, 14 Dec 2021 20:57:57 GMT" }, { "version": "v67", "created": "Mon, 10 Jan 2022 09:58:37 GMT" }, { "version": "v68", "created": "Sun, 24 Apr 2022 06:42:13 GMT" }, { "version": "v69", "created": "Tue, 23 Aug 2022 06:41:40 GMT" }, { "version": "v7", "created": "Sun, 27 May 2012 08:15:49 GMT" }, { "version": "v70", "created": "Mon, 3 Oct 2022 09:26:26 GMT" }, { "version": "v71", "created": "Thu, 10 Nov 2022 13:27:56 GMT" }, { "version": "v72", "created": "Wed, 18 Jan 2023 08:58:39 GMT" }, { "version": "v73", "created": "Mon, 13 Mar 2023 03:25:16 GMT" }, { "version": "v74", "created": "Sun, 2 Apr 2023 10:34:41 GMT" }, { "version": "v75", "created": "Mon, 1 May 2023 05:26:28 GMT" }, { "version": "v76", "created": "Sun, 4 Jun 2023 10:38:47 GMT" }, { "version": "v77", "created": "Sun, 9 Jul 2023 23:25:20 GMT" }, { "version": "v78", "created": "Mon, 21 Aug 2023 08:40:32 GMT" }, { "version": "v79", "created": "Wed, 13 Sep 2023 09:55:30 GMT" }, { "version": "v8", "created": "Wed, 15 Aug 2012 12:11:34 GMT" }, { "version": "v80", "created": "Thu, 5 Oct 2023 13:29:54 GMT" }, { "version": "v9", "created": "Wed, 29 Aug 2012 06:39:31 GMT" } ]
2023-10-06T00:00:00
[ [ "Du", "Lizhi", "" ] ]
not_new_dataset
0.997305
1912.05957
Hamid Mohammadi
Hamid Mohammadi, Seyed Hossein Khasteh
Text as Environment: A Deep Reinforcement Learning Text Readability Assessment Model
null
null
null
null
cs.CL cs.LG
http://creativecommons.org/licenses/by/4.0/
Evaluating the readability of a text can significantly facilitate the precise expression of information in written form. The formulation of text readability assessment involves the identification of meaningful properties of the text regardless of its length. Sophisticated features and models are used to evaluate the comprehensibility of texts accurately. Despite this, the problem of assessing texts' readability efficiently remains relatively untouched. The efficiency of state-of-the-art text readability assessment models can be further improved using deep reinforcement learning models. Using a hard attention-based active inference technique, the proposed approach makes efficient use of input text and computational resources. Through the use of semi-supervised signals, the reinforcement learning model uses the minimum amount of text in order to determine text's readability. A comparison of the model on Weebit and Cambridge Exams with state-of-the-art models, such as the BERT text readability model, shows that it is capable of achieving state-of-the-art accuracy with a significantly smaller amount of input text than other models.
[ { "version": "v1", "created": "Thu, 12 Dec 2019 13:54:09 GMT" }, { "version": "v2", "created": "Sun, 15 Dec 2019 15:46:55 GMT" }, { "version": "v3", "created": "Wed, 4 Oct 2023 19:09:25 GMT" } ]
2023-10-06T00:00:00
[ [ "Mohammadi", "Hamid", "" ], [ "Khasteh", "Seyed Hossein", "" ] ]
not_new_dataset
0.997235
2004.05672
Julliano Rosa Nascimento
Flavia Bonomo-Braberman, Julliano R. Nascimento, Fabiano S. Oliveira, U\'everton S. Souza, and Jayme L. Szwarcfiter
Linear-time Algorithms for Eliminating Claws in Graphs
20 pages
International Transactions in Operational Research 31 (2024), 296--315
10.1111/itor.13100
null
cs.DS
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Since many NP-complete graph problems have been shown polynomial-time solvable when restricted to claw-free graphs, we study the problem of determining the distance of a given graph to a claw-free graph, considering vertex elimination as measure. CLAW-FREE VERTEX DELETION (CFVD) consists of determining the minimum number of vertices to be removed from a graph such that the resulting graph is claw-free. Although CFVD is NP-complete in general and recognizing claw-free graphs is still a challenge, where the current best algorithm for a graph $G$ has the same running time of the best algorithm for matrix multiplication, we present linear-time algorithms for CFVD on weighted block graphs and weighted graphs with bounded treewidth. Furthermore, we show that this problem can be solved in linear time by a simpler algorithm on forests, and we determine the exact values for full $k$-ary trees. On the other hand, we show that CLAW-FREE VERTEX DELETION is NP-complete even when the input graph is a split graph. We also show that the problem is hard to approximate within any constant factor better than $2$, assuming the Unique Games Conjecture.
[ { "version": "v1", "created": "Sun, 12 Apr 2020 18:49:41 GMT" } ]
2023-10-06T00:00:00
[ [ "Bonomo-Braberman", "Flavia", "" ], [ "Nascimento", "Julliano R.", "" ], [ "Oliveira", "Fabiano S.", "" ], [ "Souza", "Uéverton S.", "" ], [ "Szwarcfiter", "Jayme L.", "" ] ]
not_new_dataset
0.997512
2010.11559
Yangjing Zhang
Yangjing Zhang, Kim-Chuan Toh, Defeng Sun
Learning Graph Laplacian with MCP
32 pages
null
null
null
cs.LG math.OC
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We consider the problem of learning a graph under the Laplacian constraint with a non-convex penalty: minimax concave penalty (MCP). For solving the MCP penalized graphical model, we design an inexact proximal difference-of-convex algorithm (DCA) and prove its convergence to critical points. We note that each subproblem of the proximal DCA enjoys the nice property that the objective function in its dual problem is continuously differentiable with a semismooth gradient. Therefore, we apply an efficient semismooth Newton method to subproblems of the proximal DCA. Numerical experiments on various synthetic and real data sets demonstrate the effectiveness of the non-convex penalty MCP in promoting sparsity. Compared with the existing state-of-the-art method, our method is demonstrated to be more efficient and reliable for learning graph Laplacian with MCP.
[ { "version": "v1", "created": "Thu, 22 Oct 2020 09:33:49 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 08:56:20 GMT" } ]
2023-10-06T00:00:00
[ [ "Zhang", "Yangjing", "" ], [ "Toh", "Kim-Chuan", "" ], [ "Sun", "Defeng", "" ] ]
not_new_dataset
0.997341
2011.15122
Willem van Jaarsveld
Tarkan Temiz\"oz, Christina Imdahl, Remco Dijkman, Douniel Lamghari-Idrissi, Willem van Jaarsveld
Deep Controlled Learning for Inventory Control
null
null
null
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
Problem Definition: Are traditional deep reinforcement learning (DRL) algorithms, developed for a broad range of purposes including game-play and robotics, the most suitable machine learning algorithms for applications in inventory control? To what extent would DRL algorithms tailored to the unique characteristics of inventory control problems provide superior performance compared to DRL and traditional benchmarks? Methodology/results: We propose and study Deep Controlled Learning (DCL), a new DRL framework based on approximate policy iteration specifically designed to tackle inventory problems. Comparative evaluations reveal that DCL outperforms existing state-of-the-art heuristics in lost sales inventory control, perishable inventory systems, and inventory systems with random lead times, achieving lower average costs across all test instances and maintaining an optimality gap of no more than 0.1\%. Notably, the same hyperparameter set is utilized across all experiments, underscoring the robustness and generalizability of the proposed method. Managerial implications: These substantial performance and robustness improvements pave the way for the effective application of tailored DRL algorithms to inventory management problems, empowering decision-makers to optimize stock levels, minimize costs, and enhance responsiveness across various industries.
[ { "version": "v1", "created": "Mon, 30 Nov 2020 18:53:08 GMT" }, { "version": "v2", "created": "Thu, 9 Sep 2021 10:08:31 GMT" }, { "version": "v3", "created": "Tue, 9 Nov 2021 14:59:21 GMT" }, { "version": "v4", "created": "Fri, 12 Nov 2021 11:47:09 GMT" }, { "version": "v5", "created": "Mon, 25 Sep 2023 08:06:08 GMT" }, { "version": "v6", "created": "Thu, 28 Sep 2023 06:37:18 GMT" } ]
2023-10-06T00:00:00
[ [ "Temizöz", "Tarkan", "" ], [ "Imdahl", "Christina", "" ], [ "Dijkman", "Remco", "" ], [ "Lamghari-Idrissi", "Douniel", "" ], [ "van Jaarsveld", "Willem", "" ] ]
not_new_dataset
0.997478
2102.00696
Selim Furkan Tekin
Selim Furkan Tekin, Arda Fazla and Suleyman Serdar Kozat
Numerical Weather Forecasting using Convolutional-LSTM with Attention and Context Matcher Mechanisms
- In our journal submission, we removed the integration of the observational data section since it was not used in the experiments. Thus, we also removed the authors from the paper who were responsible for that section. - In the second version, we also performed an experiment on WeatherBench. We compare our results with the Physical Weather Forecasting Models
null
null
null
cs.LG cs.AI cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Numerical weather forecasting using high-resolution physical models often requires extensive computational resources on supercomputers, which diminishes their wide usage in most real-life applications. As a remedy, applying deep learning methods has revealed innovative solutions within this field. To this end, we introduce a novel deep learning architecture for forecasting high-resolution spatio-temporal weather data. Our approach extends the conventional encoder-decoder structure by integrating Convolutional Long-short Term Memory and Convolutional Neural Networks. In addition, we incorporate attention and context matcher mechanisms into the model architecture. Our Weather Model achieves significant performance improvements compared to baseline deep learning models, including ConvLSTM, TrajGRU, and U-Net. Our experimental evaluation involves high-scale, real-world benchmark numerical weather datasets, namely the ERA5 hourly dataset on pressure levels and WeatherBench. Our results demonstrate substantial improvements in identifying spatial and temporal correlations with attention matrices focusing on distinct parts of the input series to model atmospheric circulations. We also compare our model with high-resolution physical models using the benchmark metrics and show that our Weather Model is accurate and easy to interpret.
[ { "version": "v1", "created": "Mon, 1 Feb 2021 08:30:42 GMT" }, { "version": "v2", "created": "Wed, 4 Oct 2023 18:56:52 GMT" } ]
2023-10-06T00:00:00
[ [ "Tekin", "Selim Furkan", "" ], [ "Fazla", "Arda", "" ], [ "Kozat", "Suleyman Serdar", "" ] ]
not_new_dataset
0.997217
2103.04904
Laszlo Csirmaz
Laszlo Csirmaz, Franti\v{s}ek Mat\'u\v{s} and Carles Padr\'o
Bipartite secret sharing and staircases
To appear in Discrete Mathematics
null
null
null
cs.CR cs.IT math.IT
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Bipartite secret sharing schemes have a bipartite access structure in which the set of participants is divided into two parts and all participants in the same part play an equivalent role. Such a bipartite scheme can be described by a \emph{staircase}: the collection of its minimal points. The complexity of a scheme is the maximal share size relative to the secret size; and the $\kappa$-complexity of an access structure is the best lower bound provided by the entropy method. An access structure is $\kappa$-ideal if it has $\kappa$-complexity 1. Motivated by the abundance of open problems in this area, the main results can be summarized as follows. First, a new characterization of $\kappa$-ideal multipartite access structures is given which offers a straightforward and simple approach to describe ideal bipartite and tripartite access structures. Second, the $\kappa$-complexity is determined for a range of bipartite access structures, including those determined by two points, staircases with equal widths and heights, and staircases with all heights 1. Third, matching linear schemes are presented for some non-ideal cases, including staircases where all heights are 1 and all widths are equal. Finally, finding the Shannon complexity of a bipartite access structure can be considered as a discrete submodular optimization problem. An interesting and intriguing continuous version is defined which might give further insight to the large-scale behavior of these optimization problems.
[ { "version": "v1", "created": "Mon, 8 Mar 2021 17:09:43 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 14:19:21 GMT" } ]
2023-10-06T00:00:00
[ [ "Csirmaz", "Laszlo", "" ], [ "Matúš", "František", "" ], [ "Padró", "Carles", "" ] ]
not_new_dataset
0.997313
2104.03937
Flavia Bonomo
Flavia Bonomo-Braberman and Gast\'on Abel Brito
Intersection models and forbidden pattern characterizations for 2-thin and proper 2-thin graphs
An extended abstract of this work, entitled "Intersection models for 2-thin and proper 2-thin graphs", was presented at LAGOS 2021 and appears in Procedia Computer Science 195 (2021), 221-229 (Proc. LAGOS'21, Sao Paulo, Brazil)
Discrete Applied Mathematics 339 (2023), 53-77
10.1016/j.dam.2023.06.013
null
cs.DM math.CO
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The \emph{thinness} of a graph is a width parameter that generalizes some properties of interval graphs, which are exactly the graphs of thinness one. Graphs with thinness at most two include, for example, bipartite convex graphs. Many NP-complete problems can be solved in polynomial time for graphs with bounded thinness, given a suitable representation of the graph. \emph{Proper thinness} is defined analogously, generalizing proper interval graphs, and a larger family of NP-complete problems are known to be polynomially solvable for graphs with bounded proper thinness. The complexity of recognizing 2-thin and proper 2-thin graphs is still open. In this work, we present characterizations of 2-thin and proper 2-thin graphs as intersection graphs of rectangles in the plane, as vertex intersection graphs of paths on a grid (VPG graphs), and by forbidden ordered patterns. We also prove that independent 2-thin graphs are exactly the interval bigraphs, and that proper independent 2-thin graphs are exactly the bipartite permutation graphs. Finally, we take a step towards placing the thinness and its variations in the landscape of width parameters, by upper bounding the proper thinness in terms of the bandwidth.
[ { "version": "v1", "created": "Thu, 8 Apr 2021 17:31:41 GMT" }, { "version": "v2", "created": "Sat, 1 Apr 2023 19:20:18 GMT" } ]
2023-10-06T00:00:00
[ [ "Bonomo-Braberman", "Flavia", "" ], [ "Brito", "Gastón Abel", "" ] ]
not_new_dataset
0.99737
2104.07454
Rohitash Chandra
Animesh Renanse, Alok Sharma, Rohitash Chandra
Memory Capacity of Recurrent Neural Networks with Matrix Representation
null
null
null
null
cs.LG cs.AI cs.CC
http://creativecommons.org/licenses/by/4.0/
It is well known that canonical recurrent neural networks (RNNs) face limitations in learning long-term dependencies which have been addressed by memory structures in long short-term memory (LSTM) networks. Neural Turing machines (NTMs) are novel RNNs that implement the notion of programmable computers with neural network controllers that can learn simple algorithmic tasks. Matrix neural networks feature matrix representation which inherently preserves the spatial structure of data when compared to canonical neural networks that use vector-based representation. One may then argue that neural networks with matrix representations may have the potential to provide better memory capacity. In this paper, we define and study a probabilistic notion of memory capacity based on Fisher information for matrix-based RNNs. We find bounds on memory capacity for such networks under various hypotheses and compare them with their vector counterparts. In particular, we show that the memory capacity of such networks is bounded by $N^2$ for $N\times N$ state matrix which generalizes the one known for vector networks. We also show and analyze the increase in memory capacity for such networks which is introduced when one exhibits an external state memory, such as NTMs. Consequently, we construct NTMs with RNN controllers with matrix-based representation of external memory, leading us to introduce Matrix NTMs. We demonstrate the performance of this class of memory networks under certain algorithmic learning tasks such as copying and recall and compare it with Matrix RNNs. We find an improvement in the performance of Matrix NTMs by the addition of external memory, in comparison to Matrix RNNs.
[ { "version": "v1", "created": "Sun, 11 Apr 2021 23:43:28 GMT" }, { "version": "v2", "created": "Sun, 30 Oct 2022 06:43:49 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 03:47:41 GMT" } ]
2023-10-06T00:00:00
[ [ "Renanse", "Animesh", "" ], [ "Sharma", "Alok", "" ], [ "Chandra", "Rohitash", "" ] ]
not_new_dataset
0.997475
2105.07099
Seyed Omid Davoudi
Omid Davoodi, Majid Komeili
Feature-Based Interpretable Reinforcement Learning based on State-Transition Models
null
null
10.1109/SMC52423.2021.9658917
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Growing concerns regarding the operational usage of AI models in the real-world has caused a surge of interest in explaining AI models' decisions to humans. Reinforcement Learning is not an exception in this regard. In this work, we propose a method for offering local explanations on risk in reinforcement learning. Our method only requires a log of previous interactions between the agent and the environment to create a state-transition model. It is designed to work on RL environments with either continuous or discrete state and action spaces. After creating the model, actions of any agent can be explained in terms of the features most influential in increasing or decreasing risk or any other desirable objective function in the locality of the agent. Through experiments, we demonstrate the effectiveness of the proposed method in providing such explanations.
[ { "version": "v1", "created": "Fri, 14 May 2021 23:43:11 GMT" } ]
2023-10-06T00:00:00
[ [ "Davoodi", "Omid", "" ], [ "Komeili", "Majid", "" ] ]
not_new_dataset
0.997394
2107.08086
Raman Goyal
Raman Goyal, Ran Wang, Mohamed Naveed Gul Mohamed, Aayushman Sharma, Suman Chakravorty
An Information-state based Approach to the Optimal Output Feedback Control of Nonlinear Systems
null
null
null
null
cs.RO cs.SY eess.SY
http://creativecommons.org/licenses/by/4.0/
This paper develops a data-based approach to the closed-loop output feedback control of nonlinear dynamical systems with a partial nonlinear observation model. We propose an information state based approach to rigorously transform the partially observed problem into a fully observed problem where the information state consists of the past several observations and control inputs. We further show the equivalence of the transformed and the initial partially observed optimal control problems and provide the conditions to solve for the deterministic optimal solution. We develop a data based generalization of the iterative Linear Quadratic Regulator (iLQR) to partially observed systems using a local linear time varying model of the information state dynamics approximated by an Autoregressive moving average (ARMA) model, that is generated using only the input-output data. This open-loop trajectory optimization solution is then used to design a local feedback control law, and the composite law then provides an optimum solution to the partially observed feedback design problem. The efficacy of the developed method is shown by controlling complex high dimensional nonlinear dynamical systems in the presence of model and sensing uncertainty.
[ { "version": "v1", "created": "Fri, 16 Jul 2021 19:21:43 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 16:28:20 GMT" } ]
2023-10-06T00:00:00
[ [ "Goyal", "Raman", "" ], [ "Wang", "Ran", "" ], [ "Mohamed", "Mohamed Naveed Gul", "" ], [ "Sharma", "Aayushman", "" ], [ "Chakravorty", "Suman", "" ] ]
not_new_dataset
0.99729
2108.05641
Jinpeng Chen
Jinpeng Chen, Haiyang Li, Xudong Zhang, Fan Zhang, Senzhang Wang, Kaimin Wei and Jiaqi Ji
SR-HetGNN:Session-based Recommendation with Heterogeneous Graph Neural Network
null
null
null
null
cs.IR cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The Session-Based Recommendation System aims to predict the user's next click based on their previous session sequence. The current studies generally learn user preferences according to the transitions of items in the user's session sequence. However, other effective information in the session sequence, such as user profiles, are largely ignored which may lead to the model unable to learn the user's specific preferences. In this paper, we propose SR-HetGNN, a novel session recommendation method that uses a heterogeneous graph neural network (HetGNN) to learn session embeddings and capture the specific preferences of anonymous users. Specifically, SR-HetGNN first constructs heterogeneous graphs containing various types of nodes according to the session sequence, which can capture the dependencies among items, users, and sessions. Second, HetGNN captures the complex transitions between items and learns the item embeddings containing user information. Finally, local and global session embeddings are combined with the attentional network to obtain the final session embedding, considering the influence of users' long and short-term preferences. SR-HetGNN is shown to be superior to the existing state-of-the-art session-based recommendation methods through extensive experiments over two real large datasets Diginetica and Tmall.
[ { "version": "v1", "created": "Thu, 12 Aug 2021 10:12:48 GMT" }, { "version": "v2", "created": "Wed, 4 Oct 2023 03:21:08 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 08:28:44 GMT" } ]
2023-10-06T00:00:00
[ [ "Chen", "Jinpeng", "" ], [ "Li", "Haiyang", "" ], [ "Zhang", "Xudong", "" ], [ "Zhang", "Fan", "" ], [ "Wang", "Senzhang", "" ], [ "Wei", "Kaimin", "" ], [ "Ji", "Jiaqi", "" ] ]
not_new_dataset
0.997301
2109.03890
Vignesh Viswanathan
Gagan Biradar, Vignesh Viswanathan, Yair Zick
Model Explanations via the Axiomatic Causal Lens
null
null
null
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Explaining the decisions of black-box models is a central theme in the study of trustworthy ML. Numerous measures have been proposed in the literature; however, none of them take an axiomatic approach to causal explainability. In this work, we propose three explanation measures which aggregate the set of all but-for causes -- a necessary and sufficient explanation -- into feature importance weights. Our first measure is a natural adaptation of Chockler and Halpern's notion of causal responsibility, whereas the other two correspond to existing game-theoretic influence measures. We present an axiomatic treatment for our proposed indices, showing that they can be uniquely characterized by a set of desirable properties. We also extend our approach to derive a new method to compute the Shapley-Shubik and Banzhaf indices for black-box model explanations. Finally, we analyze and compare the necessity and sufficiency of all our proposed explanation measures in practice using the Adult-Income dataset. Thus, our work is the first to formally bridge the gap between model explanations, game-theoretic influence, and causal analysis.
[ { "version": "v1", "created": "Wed, 8 Sep 2021 19:33:52 GMT" }, { "version": "v2", "created": "Fri, 17 Sep 2021 14:17:59 GMT" }, { "version": "v3", "created": "Mon, 31 Jan 2022 23:50:48 GMT" }, { "version": "v4", "created": "Mon, 11 Sep 2023 19:33:45 GMT" }, { "version": "v5", "created": "Wed, 27 Sep 2023 20:17:38 GMT" }, { "version": "v6", "created": "Wed, 4 Oct 2023 20:36:32 GMT" } ]
2023-10-06T00:00:00
[ [ "Biradar", "Gagan", "" ], [ "Viswanathan", "Vignesh", "" ], [ "Zick", "Yair", "" ] ]
not_new_dataset
0.997332
2109.04939
Ryo Yoshida
Ryo Yoshida, Hiroshi Noji, Yohei Oseki
Modeling Human Sentence Processing with Left-Corner Recurrent Neural Network Grammars
Accepted by EMNLP 2021
null
null
null
cs.CL
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In computational linguistics, it has been shown that hierarchical structures make language models (LMs) more human-like. However, the previous literature has been agnostic about a parsing strategy of the hierarchical models. In this paper, we investigated whether hierarchical structures make LMs more human-like, and if so, which parsing strategy is most cognitively plausible. In order to address this question, we evaluated three LMs against human reading times in Japanese with head-final left-branching structures: Long Short-Term Memory (LSTM) as a sequential model and Recurrent Neural Network Grammars (RNNGs) with top-down and left-corner parsing strategies as hierarchical models. Our computational modeling demonstrated that left-corner RNNGs outperformed top-down RNNGs and LSTM, suggesting that hierarchical and left-corner architectures are more cognitively plausible than top-down or sequential architectures. In addition, the relationships between the cognitive plausibility and (i) perplexity, (ii) parsing, and (iii) beam size will also be discussed.
[ { "version": "v1", "created": "Fri, 10 Sep 2021 15:35:00 GMT" }, { "version": "v2", "created": "Thu, 11 May 2023 02:41:41 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 10:33:42 GMT" } ]
2023-10-06T00:00:00
[ [ "Yoshida", "Ryo", "" ], [ "Noji", "Hiroshi", "" ], [ "Oseki", "Yohei", "" ] ]
not_new_dataset
0.997377
2110.03991
John Stephan
Rachid Guerraoui, Nirupam Gupta, Rafael Pinot, Sebastien Rouault, and John Stephan
Combining Differential Privacy and Byzantine Resilience in Distributed SGD
null
null
null
null
cs.LG cs.CR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Privacy and Byzantine resilience (BR) are two crucial requirements of modern-day distributed machine learning. The two concepts have been extensively studied individually but the question of how to combine them effectively remains unanswered. This paper contributes to addressing this question by studying the extent to which the distributed SGD algorithm, in the standard parameter-server architecture, can learn an accurate model despite (a) a fraction of the workers being malicious (Byzantine), and (b) the other fraction, whilst being honest, providing noisy information to the server to ensure differential privacy (DP). We first observe that the integration of standard practices in DP and BR is not straightforward. In fact, we show that many existing results on the convergence of distributed SGD under Byzantine faults, especially those relying on $(\alpha,f)$-Byzantine resilience, are rendered invalid when honest workers enforce DP. To circumvent this shortcoming, we revisit the theory of $(\alpha,f)$-BR to obtain an approximate convergence guarantee. Our analysis provides key insights on how to improve this guarantee through hyperparameter optimization. Essentially, our theoretical and empirical results show that (1) an imprudent combination of standard approaches to DP and BR might be fruitless, but (2) by carefully re-tuning the learning algorithm, we can obtain reasonable learning accuracy while simultaneously guaranteeing DP and BR.
[ { "version": "v1", "created": "Fri, 8 Oct 2021 09:23:03 GMT" }, { "version": "v2", "created": "Wed, 20 Oct 2021 14:21:46 GMT" }, { "version": "v3", "created": "Tue, 26 Oct 2021 13:37:16 GMT" }, { "version": "v4", "created": "Thu, 5 Oct 2023 09:03:58 GMT" } ]
2023-10-06T00:00:00
[ [ "Guerraoui", "Rachid", "" ], [ "Gupta", "Nirupam", "" ], [ "Pinot", "Rafael", "" ], [ "Rouault", "Sebastien", "" ], [ "Stephan", "John", "" ] ]
not_new_dataset
0.997476
2110.14883
Yang You
Shenggui Li and Hongxin Liu and Zhengda Bian and Jiarui Fang and Haichen Huang and Yuliang Liu and Boxiang Wang and Yang You
Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training
null
null
null
null
cs.LG cs.AI cs.CL cs.CV cs.DC
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The success of Transformer models has pushed the deep learning model scale to billions of parameters. Due to the limited memory resource of a single GPU, However, the best practice for choosing the optimal parallel strategy is still lacking, since it requires domain expertise in both deep learning and parallel computing. The Colossal-AI system addressed the above challenge by introducing a unified interface to scale your sequential code of model training to distributed environments. It supports parallel training methods such as data, pipeline, tensor, and sequence parallelism, as well as heterogeneous training methods integrated with zero redundancy optimizer. Compared to the baseline system, Colossal-AI can achieve up to 2.76 times training speedup on large-scale models.
[ { "version": "v1", "created": "Thu, 28 Oct 2021 04:45:55 GMT" }, { "version": "v2", "created": "Tue, 20 Sep 2022 12:54:20 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 04:09:09 GMT" } ]
2023-10-06T00:00:00
[ [ "Li", "Shenggui", "" ], [ "Liu", "Hongxin", "" ], [ "Bian", "Zhengda", "" ], [ "Fang", "Jiarui", "" ], [ "Huang", "Haichen", "" ], [ "Liu", "Yuliang", "" ], [ "Wang", "Boxiang", "" ], [ "You", "Yang", "" ] ]
not_new_dataset
0.997264
2110.15497
Peiyu Yu
Peiyu Yu, Sirui Xie, Xiaojian Ma, Yixin Zhu, Ying Nian Wu, Song-Chun Zhu
Unsupervised Foreground Extraction via Deep Region Competition
NeurIPS 2021
null
null
null
cs.CV cs.LG
http://creativecommons.org/licenses/by-sa/4.0/
We present Deep Region Competition (DRC), an algorithm designed to extract foreground objects from images in a fully unsupervised manner. Foreground extraction can be viewed as a special case of generic image segmentation that focuses on identifying and disentangling objects from the background. In this work, we rethink the foreground extraction by reconciling energy-based prior with generative image modeling in the form of Mixture of Experts (MoE), where we further introduce the learned pixel re-assignment as the essential inductive bias to capture the regularities of background regions. With this modeling, the foreground-background partition can be naturally found through Expectation-Maximization (EM). We show that the proposed method effectively exploits the interaction between the mixture components during the partitioning process, which closely connects to region competition, a seminal approach for generic image segmentation. Experiments demonstrate that DRC exhibits more competitive performances on complex real-world data and challenging multi-object scenes compared with prior methods. Moreover, we show empirically that DRC can potentially generalize to novel foreground objects even from categories unseen during training.
[ { "version": "v1", "created": "Fri, 29 Oct 2021 02:32:44 GMT" }, { "version": "v2", "created": "Tue, 9 Nov 2021 01:40:02 GMT" }, { "version": "v3", "created": "Sat, 25 Dec 2021 14:18:17 GMT" }, { "version": "v4", "created": "Wed, 4 Oct 2023 22:05:42 GMT" } ]
2023-10-06T00:00:00
[ [ "Yu", "Peiyu", "" ], [ "Xie", "Sirui", "" ], [ "Ma", "Xiaojian", "" ], [ "Zhu", "Yixin", "" ], [ "Wu", "Ying Nian", "" ], [ "Zhu", "Song-Chun", "" ] ]
not_new_dataset
0.997399
2111.02062
Pio Calderon
Pio Calderon, Alexander Soen, Marian-Andrei Rizoiu
Linking Across Data Granularity: Fitting Multivariate Hawkes Processes to Partially Interval-Censored Data
This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
null
null
null
cs.LG cs.CE
http://creativecommons.org/licenses/by-nc-sa/4.0/
The multivariate Hawkes process (MHP) is widely used for analyzing data streams that interact with each other, where events generate new events within their own dimension (via self-excitation) or across different dimensions (via cross-excitation). However, in certain applications, the timestamps of individual events in some dimensions are unobservable, and only event counts within intervals are known, referred to as partially interval-censored data. The MHP is unsuitable for handling such data since its estimation requires event timestamps. In this study, we introduce the Partial Mean Behavior Poisson (PMBP) process, a novel point process which shares parameter equivalence with the MHP and can effectively model both timestamped and interval-censored data. We demonstrate the capabilities of the PMBP process using synthetic and real-world datasets. Firstly, we illustrate that the PMBP process can approximate MHP parameters and recover the spectral radius using synthetic event histories. Next, we assess the performance of the PMBP process in predicting YouTube popularity and find that it surpasses state-of-the-art methods. Lastly, we leverage the PMBP process to gain qualitative insights from a dataset comprising daily COVID-19 case counts from multiple countries and COVID-19-related news articles. By clustering the PMBP-modeled countries, we unveil hidden interaction patterns between occurrences of COVID-19 cases and news reporting.
[ { "version": "v1", "created": "Wed, 3 Nov 2021 08:25:35 GMT" }, { "version": "v2", "created": "Mon, 7 Feb 2022 04:01:58 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 04:55:06 GMT" } ]
2023-10-06T00:00:00
[ [ "Calderon", "Pio", "" ], [ "Soen", "Alexander", "" ], [ "Rizoiu", "Marian-Andrei", "" ] ]
not_new_dataset
0.997466
2111.12232
Katsuya Hotta
Katsuya Hotta, Takuya Akashi, Shogo Tokai, Chao Zhang
PMSSC: Parallelizable multi-subset based self-expressive model for subspace clustering
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Subspace clustering methods which embrace a self-expressive model that represents each data point as a linear combination of other data points in the dataset provide powerful unsupervised learning techniques. However, when dealing with large datasets, representation of each data point by referring to all data points via a dictionary suffers from high computational complexity. To alleviate this issue, we introduce a parallelizable multi-subset based self-expressive model (PMS) which represents each data point by combining multiple subsets, with each consisting of only a small proportion of the samples. The adoption of PMS in subspace clustering (PMSSC) leads to computational advantages because the optimization problems decomposed over each subset are small, and can be solved efficiently in parallel. Furthermore, PMSSC is able to combine multiple self-expressive coefficient vectors obtained from subsets, which contributes to an improvement in self-expressiveness. Extensive experiments on synthetic and real-world datasets show the efficiency and effectiveness of our approach in comparison to other methods.
[ { "version": "v1", "created": "Wed, 24 Nov 2021 02:22:43 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 16:30:48 GMT" } ]
2023-10-06T00:00:00
[ [ "Hotta", "Katsuya", "" ], [ "Akashi", "Takuya", "" ], [ "Tokai", "Shogo", "" ], [ "Zhang", "Chao", "" ] ]
not_new_dataset
0.997267
2112.03379
Seungwoo Jeong
Seungwoo Jeong, Wonjun Ko, Ahmad Wisnu Mulyadi, Heung-Il Suk
Efficient Continuous Manifold Learning for Time Series Modeling
null
null
10.1109/TPAMI.2023.3320125
null
cs.LG cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Modeling non-Euclidean data is drawing attention along with the unprecedented successes of deep neural networks in diverse fields. In particular, symmetric positive definite (SPD) matrix is being actively studied in computer vision, signal processing, and medical image analysis, thanks to its ability to learn appropriate statistical representations. However, due to its strong constraints, it remains challenging for optimization problems or inefficient computation costs, especially, within a deep learning framework. In this paper, we propose to exploit a diffeomorphism mapping between Riemannian manifolds and a Cholesky space, by which it becomes feasible not only to efficiently solve optimization problems but also to reduce computation costs greatly. Further, in order for dynamics modeling in time series data, we devise a continuous manifold learning method by integrating a manifold ordinary differential equation and a gated recurrent neural network in a systematic manner. It is noteworthy that because of the nice parameterization of matrices in a Cholesky space, it is straightforward to train our proposed network with Riemannian geometric metrics equipped. We demonstrate through experiments that the proposed model can be efficiently and reliably trained as well as outperform existing manifold methods and state-of-the-art methods in two classification tasks: action recognition and sleep staging classification.
[ { "version": "v1", "created": "Fri, 3 Dec 2021 01:38:38 GMT" } ]
2023-10-06T00:00:00
[ [ "Jeong", "Seungwoo", "" ], [ "Ko", "Wonjun", "" ], [ "Mulyadi", "Ahmad Wisnu", "" ], [ "Suk", "Heung-Il", "" ] ]
not_new_dataset
0.997369
2112.08581
Weijie Zheng
Weijie Zheng, Benjamin Doerr
Mathematical Runtime Analysis for the Non-Dominated Sorting Genetic Algorithm II (NSGA-II)
Accepted for publication in "Artificial Intelligence". This is the journal version of the paper "Weijie Zheng, Yufei Liu, Benjamin Doerr: A First Mathematical Runtime Analysis of the Non-Dominated Sorting Genetic Algorithm II (NSGA-II). AAAI 2022. arXiv:2112.08581v3"
Artificial Intelligence 325 (2023), 104016
10.1016/j.artint.2023.104016
null
cs.NE cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The non-dominated sorting genetic algorithm II (NSGA-II) is the most intensively used multi-objective evolutionary algorithm (MOEA) in real-world applications. However, in contrast to several simple MOEAs analyzed also via mathematical means, no such study exists for the NSGA-II so far. In this work, we show that mathematical runtime analyses are feasible also for the NSGA-II. As particular results, we prove that with a population size four times larger than the size of the Pareto front, the NSGA-II with two classic mutation operators and four different ways to select the parents satisfies the same asymptotic runtime guarantees as the SEMO and GSEMO algorithms on the basic OneMinMax and LeadingOnesTrailingZeros benchmarks. However, if the population size is only equal to the size of the Pareto front, then the NSGA-II cannot efficiently compute the full Pareto front: for an exponential number of iterations, the population will always miss a constant fraction of the Pareto front. Our experiments confirm the above findings.
[ { "version": "v1", "created": "Thu, 16 Dec 2021 03:00:20 GMT" }, { "version": "v2", "created": "Fri, 11 Feb 2022 15:16:57 GMT" }, { "version": "v3", "created": "Mon, 20 Jun 2022 10:31:08 GMT" }, { "version": "v4", "created": "Fri, 24 Jun 2022 11:24:43 GMT" }, { "version": "v5", "created": "Sun, 9 Jul 2023 12:19:54 GMT" }, { "version": "v6", "created": "Mon, 18 Sep 2023 14:03:23 GMT" } ]
2023-10-06T00:00:00
[ [ "Zheng", "Weijie", "" ], [ "Doerr", "Benjamin", "" ] ]
not_new_dataset
0.997388
2202.09573
Gabriel Turinici
Gabriel Turinici
Diversity in deep generative models and generative AI
null
null
null
null
cs.CV cs.AI
http://creativecommons.org/licenses/by-nc-nd/4.0/
The decoder-based machine learning generative algorithms such as Generative Adversarial Networks (GAN), Variational Auto-Encoders (VAE), Transformers show impressive results when constructing objects similar to those in a training ensemble. However, the generation of new objects builds mainly on the understanding of the hidden structure of the training dataset followed by a sampling from a multi-dimensional normal variable. In particular each sample is independent from the others and can repeatedly propose same type of objects. To cure this drawback we introduce a kernel-based measure quantization method that can produce new objects from a given target measure by approximating it as a whole and even staying away from elements already drawn from that distribution. This ensures a better diversity of the produced objects. The method is tested on classic machine learning benchmarks.
[ { "version": "v1", "created": "Sat, 19 Feb 2022 10:52:52 GMT" }, { "version": "v2", "created": "Fri, 15 Sep 2023 16:55:40 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 13:32:57 GMT" } ]
2023-10-06T00:00:00
[ [ "Turinici", "Gabriel", "" ] ]
not_new_dataset
0.997369
2202.13103
Prerona Chatterjee
Prerona Chatterjee, Kshitij Gajjar, Anamay Tengse
Monotone Classes Beyond VNP
30 pages; made changes suggested by reviewers
null
null
null
cs.CC
http://creativecommons.org/licenses/by-nc-sa/4.0/
In this work, we study the natural monotone analogues of various equivalent definitions of VPSPACE: a well studied class (Poizat 2008, Koiran and Perifel 2009, Malod 2011, Mahajan and Rao 2013) that is believed to be larger than VNP. We observe that these monotone analogues are not equivalent unlike their non-monotone counterparts, and propose monotone VPSPACE (mVPSPACE) to be defined as the monotone analogue of Poizat's definition. With this definition, mVPSPACE turns out to be exponentially stronger than mVNP and also satisfies several desirable closure properties that the other analogues may not. Our initial goal was to understand the monotone complexity of transparent polynomials, a concept that was recently introduced by Hrube\v{s} and Yehudayoff (2021). In that context, we show that transparent polynomials of large sparsity are hard for the monotone analogues of all the known definitions of VPSPACE, except for the one due to Poizat.
[ { "version": "v1", "created": "Sat, 26 Feb 2022 10:18:15 GMT" }, { "version": "v2", "created": "Mon, 26 Sep 2022 15:17:49 GMT" }, { "version": "v3", "created": "Sun, 23 Jul 2023 12:48:01 GMT" }, { "version": "v4", "created": "Thu, 5 Oct 2023 14:21:06 GMT" } ]
2023-10-06T00:00:00
[ [ "Chatterjee", "Prerona", "" ], [ "Gajjar", "Kshitij", "" ], [ "Tengse", "Anamay", "" ] ]
not_new_dataset
0.99735
2205.05250
Hao Ren
Hao Ren, Xiaojun Liang, Chunhua Yang, Zhiwen Chen, and Weihua Gui
Spatial-temporal associations representation and application for process monitoring using graph convolution neural network
null
null
null
null
cs.LG cs.AI cs.SY eess.SY
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Thank you very much for the attention and concern of colleagues and scholars in this work. With the comments and guidance of experts, editors, and reviewers, this work has been accepted for publishing in the journal "Process Safety and Environmental Protection". The theme of this paper relies on the Spatial-temporal associations of numerous variables in the same industrial processes, which refers to numerous variables obtained in dynamic industrial processes with Spatial-temporal correlation characteristics, i.e., these variables are not only highly correlated in time but also interrelated in space. To handle this problem, three key issues need to be well addressed: variable characteristics modeling and representation, graph network construction (temporal information), and graph characteristics perception. The first issue is implemented by assuming the data follows one improved Gaussian distribution, while the graph network can be defined by the monitoring variables and their edges which are calculated by their characteristics in time. Finally, these networks corresponding to process states at different times are fed into a graph convolutional neural network to implement graph classification to achieve process monitoring. A benchmark experiment (Tennessee Eastman chemical process) and one application study (cobalt purification from zinc solution) are employed to demonstrate the feasibility and applicability of this paper.
[ { "version": "v1", "created": "Wed, 11 May 2022 03:36:35 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 14:32:15 GMT" } ]
2023-10-06T00:00:00
[ [ "Ren", "Hao", "" ], [ "Liang", "Xiaojun", "" ], [ "Yang", "Chunhua", "" ], [ "Chen", "Zhiwen", "" ], [ "Gui", "Weihua", "" ] ]
not_new_dataset
0.997368
2205.09174
Ehud Shapiro
Idit Keidar, Oded Naor, Ouri Poupko, and Ehud Shapiro
Cordial Miners: Fast and Efficient Consensus for Every Eventuality
null
null
10.4230/LIPIcs.DISC.2023.26
null
cs.DC cs.MA cs.NI
http://creativecommons.org/licenses/by-nc-nd/4.0/
Cordial Miners are a family of efficient Byzantine Atomic Broadcast protocols, with instances for asynchrony and eventual synchrony. They improve the latency of state-of-the-art DAG-based protocols by almost 2X and achieve optimal good-case complexity of O(n) by forgoing Reliable Broadcast as a building block. Rather, Cordial Miners use the blocklace -- a partially-ordered counterpart of the totally-ordered blockchain data structure -- to implement the three algorithmic components of consensus: Dissemination, equivocation-exclusion, and ordering.
[ { "version": "v1", "created": "Wed, 18 May 2022 18:45:20 GMT" }, { "version": "v2", "created": "Mon, 1 Aug 2022 02:28:16 GMT" }, { "version": "v3", "created": "Thu, 11 Aug 2022 19:19:31 GMT" }, { "version": "v4", "created": "Wed, 9 Nov 2022 19:34:18 GMT" }, { "version": "v5", "created": "Thu, 11 May 2023 17:39:43 GMT" }, { "version": "v6", "created": "Fri, 22 Sep 2023 20:40:09 GMT" } ]
2023-10-06T00:00:00
[ [ "Keidar", "Idit", "" ], [ "Naor", "Oded", "" ], [ "Poupko", "Ouri", "" ], [ "Shapiro", "Ehud", "" ] ]
not_new_dataset
0.996969
2206.05895
Peiyu Yu
Peiyu Yu, Sirui Xie, Xiaojian Ma, Baoxiong Jia, Bo Pang, Ruiqi Gao, Yixin Zhu, Song-Chun Zhu, and Ying Nian Wu
Latent Diffusion Energy-Based Model for Interpretable Text Modeling
ICML 2022
null
null
null
cs.LG cs.CL
http://creativecommons.org/licenses/by-nc-sa/4.0/
Latent space Energy-Based Models (EBMs), also known as energy-based priors, have drawn growing interests in generative modeling. Fueled by its flexibility in the formulation and strong modeling power of the latent space, recent works built upon it have made interesting attempts aiming at the interpretability of text modeling. However, latent space EBMs also inherit some flaws from EBMs in data space; the degenerate MCMC sampling quality in practice can lead to poor generation quality and instability in training, especially on data with complex latent structures. Inspired by the recent efforts that leverage diffusion recovery likelihood learning as a cure for the sampling issue, we introduce a novel symbiosis between the diffusion models and latent space EBMs in a variational learning framework, coined as the latent diffusion energy-based model. We develop a geometric clustering-based regularization jointly with the information bottleneck to further improve the quality of the learned latent space. Experiments on several challenging tasks demonstrate the superior performance of our model on interpretable text modeling over strong counterparts.
[ { "version": "v1", "created": "Mon, 13 Jun 2022 03:41:31 GMT" }, { "version": "v2", "created": "Tue, 14 Jun 2022 03:01:05 GMT" }, { "version": "v3", "created": "Mon, 4 Jul 2022 16:28:58 GMT" }, { "version": "v4", "created": "Wed, 4 Oct 2023 22:00:21 GMT" } ]
2023-10-06T00:00:00
[ [ "Yu", "Peiyu", "" ], [ "Xie", "Sirui", "" ], [ "Ma", "Xiaojian", "" ], [ "Jia", "Baoxiong", "" ], [ "Pang", "Bo", "" ], [ "Gao", "Ruiqi", "" ], [ "Zhu", "Yixin", "" ], [ "Zhu", "Song-Chun", "" ], [ "Wu", "Ying Nian", "" ] ]
not_new_dataset
0.997415
2207.03299
Juan Bascur
Juan Pablo Bascur, Suzan Verberne, Nees Jan van Eck, Ludo Waltman
Academic information retrieval using citation clusters: In-depth evaluation based on systematic reviews
Final version
null
null
null
cs.DL
http://creativecommons.org/licenses/by/4.0/
The field of scientometrics has shown the power of citation-based clusters for literature analysis, yet this technique has barely been used for information retrieval tasks. This work evaluates the performance of citation based-clusters for information retrieval tasks. We simulated a search process using these clusters with a tree hierarchy of clusters and a cluster selection algorithm. We evaluated the task of finding the relevant documents for 25 systematic reviews. Our evaluation considered several trade-offs between recall and precision for the cluster selection, and we also replicated the Boolean queries self-reported by the systematic review to serve as a reference. We found that citation-based clusters search performance is highly variable and unpredictable, that it works best for users that prefer recall over precision at a ratio between 2 and 8, and that when used along with query-based search they complement each other, including finding new relevant documents.
[ { "version": "v1", "created": "Thu, 7 Jul 2022 13:50:27 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 15:42:11 GMT" } ]
2023-10-06T00:00:00
[ [ "Bascur", "Juan Pablo", "" ], [ "Verberne", "Suzan", "" ], [ "van Eck", "Nees Jan", "" ], [ "Waltman", "Ludo", "" ] ]
not_new_dataset
0.997488
2207.05132
Arghavan Moradi Dakhel
Arghavan Moradi Dakhel, Michel C. Desmarais, Foutse Khomh
Dev2vec: Representing Domain Expertise of Developers in an Embedding Space
30 pages, 5 figures
null
10.1016/j.infsof.2023.107218
null
cs.SE cs.IR cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Accurate assessment of the domain expertise of developers is important for assigning the proper candidate to contribute to a project or to attend a job role. Since the potential candidate can come from a large pool, the automated assessment of this domain expertise is a desirable goal. While previous methods have had some success within a single software project, the assessment of a developer's domain expertise from contributions across multiple projects is more challenging. In this paper, we employ doc2vec to represent the domain expertise of developers as embedding vectors. These vectors are derived from different sources that contain evidence of developers' expertise, such as the description of repositories that they contributed, their issue resolving history, and API calls in their commits. We name it dev2vec and demonstrate its effectiveness in representing the technical specialization of developers. Our results indicate that encoding the expertise of developers in an embedding vector outperforms state-of-the-art methods and improves the F1-score up to 21%. Moreover, our findings suggest that ``issue resolving history'' of developers is the most informative source of information to represent the domain expertise of developers in embedding spaces.
[ { "version": "v1", "created": "Mon, 11 Jul 2022 18:56:49 GMT" } ]
2023-10-06T00:00:00
[ [ "Dakhel", "Arghavan Moradi", "" ], [ "Desmarais", "Michel C.", "" ], [ "Khomh", "Foutse", "" ] ]
not_new_dataset
0.996237
2207.11447
Xu Zhou
Xu Zhou, Xinyu Lei, Cong Yang, Yichun Shi, Xiao Zhang, Jingwen Shi
Handling Data Heterogeneity in Federated Learning via Knowledge Distillation and Fusion
15 pages, 3 figures
null
null
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Federated learning (FL) supports distributed training of a global machine learning model across multiple devices with the help of a central server. However, data heterogeneity across different devices leads to the client model drift issue and results in model performance degradation and poor model fairness. To address the issue, we design Federated learning with global-local Knowledge Fusion (FedKF) scheme in this paper. The key idea in FedKF is to let the server return the global knowledge to be fused with the local knowledge in each training round so that the local model can be regularized towards the global optima. Therefore, the client model drift issue can be mitigated. In FedKF, we first propose the active-inactive model aggregation technique that supports a precise global knowledge representation. Then, we propose a data-free knowledge distillation (KD) approach to enable each client model to learn the global knowledge (embedded in the global model) while each client model can still learn the local knowledge (embedded in the local dataset) simultaneously, thereby realizing the global-local knowledge fusion process. The theoretical analysis and intensive experiments demonstrate the superiority of FedKF over previous solutions.
[ { "version": "v1", "created": "Sat, 23 Jul 2022 07:20:22 GMT" }, { "version": "v2", "created": "Wed, 4 Oct 2023 20:44:04 GMT" } ]
2023-10-06T00:00:00
[ [ "Zhou", "Xu", "" ], [ "Lei", "Xinyu", "" ], [ "Yang", "Cong", "" ], [ "Shi", "Yichun", "" ], [ "Zhang", "Xiao", "" ], [ "Shi", "Jingwen", "" ] ]
not_new_dataset
0.997376
2207.11880
Huaxiong Li
Kaiyi Luo, Chao Zhang, Huaxiong Li, Xiuyi Jia, Chunlin Chen
Adaptive Marginalized Semantic Hashing for Unpaired Cross-Modal Retrieval
null
null
10.1109/TMM.2023.3245400
null
cs.MM
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In recent years, Cross-Modal Hashing (CMH) has aroused much attention due to its fast query speed and efficient storage. Previous literatures have achieved promising results for Cross-Modal Retrieval (CMR) by discovering discriminative hash codes and modality-specific hash functions. Nonetheless, most existing CMR works are subjected to some restrictions: 1) It is assumed that data of different modalities are fully paired, which is impractical in real applications due to sample missing and false data alignment, and 2) binary regression targets including the label matrix and binary codes are too rigid to effectively learn semantic-preserving hash codes and hash functions. To address these problems, this paper proposes an Adaptive Marginalized Semantic Hashing (AMSH) method which not only enhances the discrimination of latent representations and hash codes by adaptive margins, but also can be used for both paired and unpaired CMR. As a two-step method, in the first step, AMSH generates semantic-aware modality-specific latent representations with adaptively marginalized labels, which enlarges the distances between different classes, and exploits the labels to preserve the inter-modal and intra-modal semantic similarities into latent representations and hash codes. In the second step, adaptive margin matrices are embedded into the hash codes, and enlarge the gaps between positive and negative bits, which improves the discrimination and robustness of hash functions. On this basis, AMSH generates similarity-preserving hash codes and robust hash functions without strict one-to-one data correspondence requirement. Experiments are conducted on several benchmark datasets to demonstrate the superiority and flexibility of AMSH over some state-of-the-art CMR methods.
[ { "version": "v1", "created": "Mon, 25 Jul 2022 02:50:20 GMT" } ]
2023-10-06T00:00:00
[ [ "Luo", "Kaiyi", "" ], [ "Zhang", "Chao", "" ], [ "Li", "Huaxiong", "" ], [ "Jia", "Xiuyi", "" ], [ "Chen", "Chunlin", "" ] ]
not_new_dataset
0.997456
2207.14096
Shaun Yuan
Gong Cheng, Xiang Yuan, Xiwen Yao, Kebing Yan, Qinghua Zeng, Xingxing Xie, and Junwei Han
Towards Large-Scale Small Object Detection: Survey and Benchmarks
in IEEE Transactions on Pattern Analysis and Machine Intelligence (2023)
IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 45, no. 11, pp. 13467-13488, 1 Nov. 2023
10.1109/TPAMI.2023.3290594
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
With the rise of deep convolutional neural networks, object detection has achieved prominent advances in past years. However, such prosperity could not camouflage the unsatisfactory situation of Small Object Detection (SOD), one of the notoriously challenging tasks in computer vision, owing to the poor visual appearance and noisy representation caused by the intrinsic structure of small targets. In addition, large-scale dataset for benchmarking small object detection methods remains a bottleneck. In this paper, we first conduct a thorough review of small object detection. Then, to catalyze the development of SOD, we construct two large-scale Small Object Detection dAtasets (SODA), SODA-D and SODA-A, which focus on the Driving and Aerial scenarios respectively. SODA-D includes 24828 high-quality traffic images and 278433 instances of nine categories. For SODA-A, we harvest 2513 high resolution aerial images and annotate 872069 instances over nine classes. The proposed datasets, as we know, are the first-ever attempt to large-scale benchmarks with a vast collection of exhaustively annotated instances tailored for multi-category SOD. Finally, we evaluate the performance of mainstream methods on SODA. We expect the released benchmarks could facilitate the development of SOD and spawn more breakthroughs in this field. Datasets and codes are available at: \url{https://shaunyuan22.github.io/SODA}.
[ { "version": "v1", "created": "Thu, 28 Jul 2022 14:02:18 GMT" }, { "version": "v2", "created": "Sun, 31 Jul 2022 08:33:25 GMT" }, { "version": "v3", "created": "Sat, 24 Dec 2022 15:43:44 GMT" }, { "version": "v4", "created": "Tue, 11 Apr 2023 03:58:28 GMT" } ]
2023-10-06T00:00:00
[ [ "Cheng", "Gong", "" ], [ "Yuan", "Xiang", "" ], [ "Yao", "Xiwen", "" ], [ "Yan", "Kebing", "" ], [ "Zeng", "Qinghua", "" ], [ "Xie", "Xingxing", "" ], [ "Han", "Junwei", "" ] ]
new_dataset
0.997998
2208.07708
Gyanendra Kumar Verma
Gyanendra K. Verma, Astha Agrawal, R. K. Sharma
Construction Methods for Galois LCD codes over Finite Fields
There are many typos and mathematical typos as well
null
null
null
cs.IT math.IT
http://creativecommons.org/licenses/by/4.0/
In this article, first we present a method for constructing many Hermitian LCD codes from a given Hermitian LCD code, and then provide several methods which utilize either a given [n, k, d] linear code or a given [n, k, d] Galois LCD code to construct new Galois LCD codes with different parameters. Using these construction methods, we construct several new [n, k, d] ternary LCD codes with better parameters for $26\leq n \leq 40$, and $21 \leq k \leq 30$. Also, optimal 2-Galois LCD codes over $\mathbb{F}_{2^3}$ for code length, $1 \leq n \leq 15$ have been obtained. Finally, we extend some previously known results to the $\sigma$-inner product from Euclidean inner product.
[ { "version": "v1", "created": "Tue, 16 Aug 2022 12:19:42 GMT" }, { "version": "v2", "created": "Fri, 2 Sep 2022 05:40:26 GMT" }, { "version": "v3", "created": "Tue, 3 Oct 2023 04:59:01 GMT" }, { "version": "v4", "created": "Thu, 5 Oct 2023 09:06:17 GMT" } ]
2023-10-06T00:00:00
[ [ "Verma", "Gyanendra K.", "" ], [ "Agrawal", "Astha", "" ], [ "Sharma", "R. K.", "" ] ]
not_new_dataset
0.997271
2209.05917
Sunkyung Lee
Eunseong Choi, Sunkyung Lee, Minjin Choi, Hyeseon Ko, Young-In Song and Jongwuk Lee
SpaDE: Improving Sparse Representations using a Dual Document Encoder for First-stage Retrieval
In Proceedings of the 31st ACM International Conference on Information and Knowledge Management (CIKM '22). 13 pages
null
10.1145/3511808.3557456
null
cs.IR
http://creativecommons.org/licenses/by-nc-sa/4.0/
Sparse document representations have been widely used to retrieve relevant documents via exact lexical matching. Owing to the pre-computed inverted index, it supports fast ad-hoc search but incurs the vocabulary mismatch problem. Although recent neural ranking models using pre-trained language models can address this problem, they usually require expensive query inference costs, implying the trade-off between effectiveness and efficiency. Tackling the trade-off, we propose a novel uni-encoder ranking model, Sparse retriever using a Dual document Encoder (SpaDE), learning document representation via the dual encoder. Each encoder plays a central role in (i) adjusting the importance of terms to improve lexical matching and (ii) expanding additional terms to support semantic matching. Furthermore, our co-training strategy trains the dual encoder effectively and avoids unnecessary intervention in training each other. Experimental results on several benchmarks show that SpaDE outperforms existing uni-encoder ranking models.
[ { "version": "v1", "created": "Tue, 13 Sep 2022 12:06:01 GMT" }, { "version": "v2", "created": "Thu, 13 Apr 2023 05:57:34 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 02:33:49 GMT" } ]
2023-10-06T00:00:00
[ [ "Choi", "Eunseong", "" ], [ "Lee", "Sunkyung", "" ], [ "Choi", "Minjin", "" ], [ "Ko", "Hyeseon", "" ], [ "Song", "Young-In", "" ], [ "Lee", "Jongwuk", "" ] ]
not_new_dataset
0.997335
2209.12148
Radu Tudor Ionescu
Neelu Madan, Nicolae-Catalin Ristea, Radu Tudor Ionescu, Kamal Nasrollahi, Fahad Shahbaz Khan, Thomas B. Moeslund, Mubarak Shah
Self-Supervised Masked Convolutional Transformer Block for Anomaly Detection
Accepted in IEEE Transactions on Pattern Analysis and Machine Intelligence
null
10.1109/TPAMI.2023.3322604
null
cs.CV cs.AI cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Anomaly detection has recently gained increasing attention in the field of computer vision, likely due to its broad set of applications ranging from product fault detection on industrial production lines and impending event detection in video surveillance to finding lesions in medical scans. Regardless of the domain, anomaly detection is typically framed as a one-class classification task, where the learning is conducted on normal examples only. An entire family of successful anomaly detection methods is based on learning to reconstruct masked normal inputs (e.g. patches, future frames, etc.) and exerting the magnitude of the reconstruction error as an indicator for the abnormality level. Unlike other reconstruction-based methods, we present a novel self-supervised masked convolutional transformer block (SSMCTB) that comprises the reconstruction-based functionality at a core architectural level. The proposed self-supervised block is extremely flexible, enabling information masking at any layer of a neural network and being compatible with a wide range of neural architectures. In this work, we extend our previous self-supervised predictive convolutional attentive block (SSPCAB) with a 3D masked convolutional layer, a transformer for channel-wise attention, as well as a novel self-supervised objective based on Huber loss. Furthermore, we show that our block is applicable to a wider variety of tasks, adding anomaly detection in medical images and thermal videos to the previously considered tasks based on RGB images and surveillance videos. We exhibit the generality and flexibility of SSMCTB by integrating it into multiple state-of-the-art neural models for anomaly detection, bringing forth empirical results that confirm considerable performance improvements on five benchmarks. We release our code and data as open source at: https://github.com/ristea/ssmctb.
[ { "version": "v1", "created": "Sun, 25 Sep 2022 04:56:10 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 10:37:39 GMT" } ]
2023-10-06T00:00:00
[ [ "Madan", "Neelu", "" ], [ "Ristea", "Nicolae-Catalin", "" ], [ "Ionescu", "Radu Tudor", "" ], [ "Nasrollahi", "Kamal", "" ], [ "Khan", "Fahad Shahbaz", "" ], [ "Moeslund", "Thomas B.", "" ], [ "Shah", "Mubarak", "" ] ]
not_new_dataset
0.988627
2210.01422
Rasool Fakoor
Rasool Fakoor and Jonas Mueller and Zachary C. Lipton and Pratik Chaudhari and Alexander J. Smola
Time-Varying Propensity Score to Bridge the Gap between the Past and Present
null
null
null
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Real-world deployment of machine learning models is challenging because data evolves over time. While no model can work when data evolves in an arbitrary fashion, if there is some pattern to these changes, we might be able to design methods to address it. This paper addresses situations when data evolves gradually. We introduce a time-varying propensity score that can detect gradual shifts in the distribution of data which allows us to selectively sample past data to update the model -- not just similar data from the past like that of a standard propensity score but also data that evolved in a similar fashion in the past. The time-varying propensity score is quite general: we demonstrate different ways of implementing it and evaluate it on a variety of problems ranging from supervised learning (e.g., image classification problems) where data undergoes a sequence of gradual shifts, to reinforcement learning tasks (e.g., robotic manipulation and continuous control) where data shifts as the policy or the task changes.
[ { "version": "v1", "created": "Tue, 4 Oct 2022 07:21:49 GMT" }, { "version": "v2", "created": "Mon, 30 Jan 2023 17:52:10 GMT" }, { "version": "v3", "created": "Wed, 14 Jun 2023 17:47:50 GMT" }, { "version": "v4", "created": "Thu, 5 Oct 2023 17:38:13 GMT" } ]
2023-10-06T00:00:00
[ [ "Fakoor", "Rasool", "" ], [ "Mueller", "Jonas", "" ], [ "Lipton", "Zachary C.", "" ], [ "Chaudhari", "Pratik", "" ], [ "Smola", "Alexander J.", "" ] ]
not_new_dataset
0.997504
2210.01944
Sajad Darabi
Sajad Darabi, Piotr Bigaj, Dawid Majchrowski, Artur Kasymov, Pawel Morkisz, Alex Fit-Florea
A Framework for Large Scale Synthetic Graph Dataset Generation
null
null
null
null
cs.LG cs.SI
http://creativecommons.org/licenses/by/4.0/
Recently there has been increasing interest in developing and deploying deep graph learning algorithms for many tasks, such as fraud detection and recommender systems. Albeit, there is a limited number of publicly available graph-structured datasets, most of which are tiny compared to production-sized applications or are limited in their application domain. This work tackles this shortcoming by proposing a scalable synthetic graph generation tool to scale the datasets to production-size graphs with trillions of edges and billions of nodes. The tool learns a series of parametric models from proprietary datasets that can be released to researchers to study various graph methods on the synthetic data increasing prototype development and novel applications. We demonstrate the generalizability of the framework across a series of datasets, mimicking structural and feature distributions as well as the ability to scale them across varying sizes demonstrating their usefulness for benchmarking and model development. Code can be found on https://github.com/NVIDIA/DeepLearningExamples/tree/master/Tools/DGLPyTorch/SyntheticGraphGeneration.
[ { "version": "v1", "created": "Tue, 4 Oct 2022 22:41:33 GMT" }, { "version": "v2", "created": "Thu, 6 Oct 2022 15:17:02 GMT" }, { "version": "v3", "created": "Tue, 28 Feb 2023 23:05:44 GMT" }, { "version": "v4", "created": "Thu, 5 Oct 2023 05:22:43 GMT" } ]
2023-10-06T00:00:00
[ [ "Darabi", "Sajad", "" ], [ "Bigaj", "Piotr", "" ], [ "Majchrowski", "Dawid", "" ], [ "Kasymov", "Artur", "" ], [ "Morkisz", "Pawel", "" ], [ "Fit-Florea", "Alex", "" ] ]
not_new_dataset
0.997341
2210.17505
Roberto Casadei PhD
Roberto Casadei, Stefano Mariani, Danilo Pianini, Mirko Viroli, Franco Zambonelli
Space-Fluid Adaptive Sampling by Self-Organisation
33 pages, 16 figures
null
null
null
cs.DC cs.AI cs.MA cs.SY eess.SY
http://creativecommons.org/licenses/by/4.0/
A recurrent task in coordinated systems is managing (estimating, predicting, or controlling) signals that vary in space, such as distributed sensed data or computation outcomes. Especially in large-scale settings, the problem can be addressed through decentralised and situated computing systems: nodes can locally sense, process, and act upon signals, and coordinate with neighbours to implement collective strategies. Accordingly, in this work we devise distributed coordination strategies for the estimation of a spatial phenomenon through collaborative adaptive sampling. Our design is based on the idea of dynamically partitioning space into regions that compete and grow/shrink to provide accurate aggregate sampling. Such regions hence define a sort of virtualised space that is "fluid", since its structure adapts in response to pressure forces exerted by the underlying phenomenon. We provide an adaptive sampling algorithm in the field-based coordination framework, and prove it is self-stabilising and locally optimal. Finally, we verify by simulation that the proposed algorithm effectively carries out a spatially adaptive sampling while maintaining a tuneable trade-off between accuracy and efficiency.
[ { "version": "v1", "created": "Mon, 31 Oct 2022 17:29:41 GMT" }, { "version": "v2", "created": "Thu, 16 Mar 2023 17:31:22 GMT" }, { "version": "v3", "created": "Tue, 1 Aug 2023 07:38:51 GMT" }, { "version": "v4", "created": "Thu, 5 Oct 2023 10:46:56 GMT" } ]
2023-10-06T00:00:00
[ [ "Casadei", "Roberto", "" ], [ "Mariani", "Stefano", "" ], [ "Pianini", "Danilo", "" ], [ "Viroli", "Mirko", "" ], [ "Zambonelli", "Franco", "" ] ]
not_new_dataset
0.997303
2211.00635
Yihan Wang
Yihan Wang, Si Si, Daliang Li, Michal Lukasik, Felix Yu, Cho-Jui Hsieh, Inderjit S Dhillon, Sanjiv Kumar
Two-stage LLM Fine-tuning with Less Specialization and More Generalization
null
null
null
null
cs.CL cs.LG
http://creativecommons.org/licenses/by/4.0/
Pretrained large language models (LLMs) are general purpose problem solvers applicable to a diverse set of tasks with prompts. They can be further improved towards a specific task by fine-tuning on a specialized dataset. However, fine-tuning usually makes the model narrowly specialized on this dataset with reduced general in-context learning performances, which is undesirable whenever the fine-tuned model needs to handle additional tasks where no fine-tuning data is available. In this work, we first demonstrate that fine-tuning on a single task indeed decreases LLMs' general in-context learning performance. We discover one important cause of such forgetting, format specialization, where the model overfits to the format of the fine-tuned task. We further show that format specialization happens at the very beginning of fine-tuning. To solve this problem, we propose Prompt Tuning with MOdel Tuning (ProMoT), a simple yet effective two-stage fine-tuning framework that reduces format specialization and improves generalization. ProMoT offloads task-specific format learning into additional and removable parameters by first doing prompt tuning and then fine-tuning the model itself with this soft prompt attached. With experiments on several fine-tuning tasks and 8 in-context evaluation tasks, we show that ProMoT achieves comparable performance on fine-tuned tasks to standard fine-tuning, but with much less loss of in-context learning performances across a board range of out-of-domain evaluation tasks. More importantly, ProMoT can even enhance generalization on in-context learning tasks that are semantically related to the fine-tuned task, e.g. ProMoT on En-Fr translation significantly improves performance on other language pairs, and ProMoT on NLI improves performance on summarization. Experiments also show that ProMoT can improve the generalization performance of multi-task training.
[ { "version": "v1", "created": "Tue, 1 Nov 2022 17:56:57 GMT" }, { "version": "v2", "created": "Wed, 4 Oct 2023 20:27:57 GMT" } ]
2023-10-06T00:00:00
[ [ "Wang", "Yihan", "" ], [ "Si", "Si", "" ], [ "Li", "Daliang", "" ], [ "Lukasik", "Michal", "" ], [ "Yu", "Felix", "" ], [ "Hsieh", "Cho-Jui", "" ], [ "Dhillon", "Inderjit S", "" ], [ "Kumar", "Sanjiv", "" ] ]
not_new_dataset
0.997431
2211.01856
Shihan Ma
Shihan Ma, Alexander Kenneth Clarke, Kostiantyn Maksymenko, Samuel Deslauriers-Gauthier, Xinjun Sheng, Xiangyang Zhu, Dario Farina
Conditional Generative Models for Simulation of EMG During Naturalistic Movements
null
null
null
null
cs.LG cs.CE eess.SP physics.bio-ph
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Numerical models of electromyographic (EMG) signals have provided a huge contribution to our fundamental understanding of human neurophysiology and remain a central pillar of motor neuroscience and the development of human-machine interfaces. However, whilst modern biophysical simulations based on finite element methods are highly accurate, they are extremely computationally expensive and thus are generally limited to modelling static systems such as isometrically contracting limbs. As a solution to this problem, we propose a transfer learning approach, in which a conditional generative model is trained to mimic the output of an advanced numerical model. To this end, we present BioMime, a conditional generative neural network trained adversarially to generate motor unit activation potential waveforms under a wide variety of volume conductor parameters. We demonstrate the ability of such a model to predictively interpolate between a much smaller number of numerical model's outputs with a high accuracy. Consequently, the computational load is dramatically reduced, which allows the rapid simulation of EMG signals during truly dynamic and naturalistic movements.
[ { "version": "v1", "created": "Thu, 3 Nov 2022 14:49:02 GMT" }, { "version": "v2", "created": "Wed, 15 Feb 2023 15:29:54 GMT" }, { "version": "v3", "created": "Thu, 13 Jul 2023 16:07:33 GMT" }, { "version": "v4", "created": "Thu, 5 Oct 2023 17:26:48 GMT" } ]
2023-10-06T00:00:00
[ [ "Ma", "Shihan", "" ], [ "Clarke", "Alexander Kenneth", "" ], [ "Maksymenko", "Kostiantyn", "" ], [ "Deslauriers-Gauthier", "Samuel", "" ], [ "Sheng", "Xinjun", "" ], [ "Zhu", "Xiangyang", "" ], [ "Farina", "Dario", "" ] ]
not_new_dataset
0.997392
2211.03660
Jiawang Bian
Libo Sun, Jia-Wang Bian, Huangying Zhan, Wei Yin, Ian Reid, Chunhua Shen
SC-DepthV3: Robust Self-supervised Monocular Depth Estimation for Dynamic Scenes
Accepted for publication in TPAMI; The code will be available at https://github.com/JiawangBian/sc_depth_pl
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Self-supervised monocular depth estimation has shown impressive results in static scenes. It relies on the multi-view consistency assumption for training networks, however, that is violated in dynamic object regions and occlusions. Consequently, existing methods show poor accuracy in dynamic scenes, and the estimated depth map is blurred at object boundaries because they are usually occluded in other training views. In this paper, we propose SC-DepthV3 for addressing the challenges. Specifically, we introduce an external pretrained monocular depth estimation model for generating single-image depth prior, namely pseudo-depth, based on which we propose novel losses to boost self-supervised training. As a result, our model can predict sharp and accurate depth maps, even when training from monocular videos of highly-dynamic scenes. We demonstrate the significantly superior performance of our method over previous methods on six challenging datasets, and we provide detailed ablation studies for the proposed terms. Source code and data will be released at https://github.com/JiawangBian/sc_depth_pl
[ { "version": "v1", "created": "Mon, 7 Nov 2022 16:17:47 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 08:53:01 GMT" } ]
2023-10-06T00:00:00
[ [ "Sun", "Libo", "" ], [ "Bian", "Jia-Wang", "" ], [ "Zhan", "Huangying", "" ], [ "Yin", "Wei", "" ], [ "Reid", "Ian", "" ], [ "Shen", "Chunhua", "" ] ]
not_new_dataset
0.997263
2211.07091
Yefei He
Yefei He, Zhenyu Lou, Luoming Zhang, Jing Liu, Weijia Wu, Hong Zhou, Bohan Zhuang
BiViT: Extremely Compressed Binary Vision Transformer
Accepted by ICCV 2023
null
null
null
cs.CV cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Model binarization can significantly compress model size, reduce energy consumption, and accelerate inference through efficient bit-wise operations. Although binarizing convolutional neural networks have been extensively studied, there is little work on exploring binarization of vision Transformers which underpin most recent breakthroughs in visual recognition. To this end, we propose to solve two fundamental challenges to push the horizon of Binary Vision Transformers (BiViT). First, the traditional binary method does not take the long-tailed distribution of softmax attention into consideration, bringing large binarization errors in the attention module. To solve this, we propose Softmax-aware Binarization, which dynamically adapts to the data distribution and reduces the error caused by binarization. Second, to better preserve the information of the pretrained model and restore accuracy, we propose a Cross-layer Binarization scheme that decouples the binarization of self-attention and multi-layer perceptrons (MLPs), and Parameterized Weight Scales which introduce learnable scaling factors for weight binarization. Overall, our method performs favorably against state-of-the-arts by 19.8% on the TinyImageNet dataset. On ImageNet, our BiViT achieves a competitive 75.6% Top-1 accuracy over Swin-S model. Additionally, on COCO object detection, our method achieves an mAP of 40.8 with a Swin-T backbone over Cascade Mask R-CNN framework.
[ { "version": "v1", "created": "Mon, 14 Nov 2022 03:36:38 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 07:59:22 GMT" } ]
2023-10-06T00:00:00
[ [ "He", "Yefei", "" ], [ "Lou", "Zhenyu", "" ], [ "Zhang", "Luoming", "" ], [ "Liu", "Jing", "" ], [ "Wu", "Weijia", "" ], [ "Zhou", "Hong", "" ], [ "Zhuang", "Bohan", "" ] ]
not_new_dataset
0.997118
2211.11961
Arghya Chakraborty
Arghya Chakraborty, Rahul Vaze
Online facility location with timed-requests and congestion
32 pages, 6 figures
null
null
null
cs.DS
http://creativecommons.org/licenses/by/4.0/
The classic online facility location problem deals with finding the optimal set of facilities in an online fashion when demand requests arrive one at a time and facilities need to be opened to service these requests. In this work, we study two variants of the online facility location problem; (1) weighted requests and (2) congestion. Both of these variants are motivated by their applications to real life scenarios and the previously known results on online facility location cannot be directly adapted to analyse them. Weighted requests: In this variant, each demand request is a pair $(x,w)$ where $x$ is the standard location of the demand while $w$ is the corresponding weight of the request. The cost of servicing request $(x,w)$ at facility $F$ is $w\cdot d(x,F)$. For this variant, given $n$ requests, we present an online algorithm attaining a competitive ratio of $\mathcal{O}(\log n)$ in the secretarial model for the weighted requests and show that it is optimal. Congestion: The congestion variant considers the case when there is an additional congestion cost that grows with the number of requests served by each facility. For this variant, when the congestion cost is a monomial, we show that there exists an algorithm attaining a constant competitive ratio. This constant is a function of the exponent of the monomial and the facility opening cost but independent of the number of requests.
[ { "version": "v1", "created": "Tue, 22 Nov 2022 02:50:51 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 15:49:18 GMT" } ]
2023-10-06T00:00:00
[ [ "Chakraborty", "Arghya", "" ], [ "Vaze", "Rahul", "" ] ]
not_new_dataset
0.997388
2211.13118
Vianney Copp\'e
Vianney Copp\'e, Xavier Gillard, Pierre Schaus
Decision Diagram-Based Branch-and-Bound with Caching for Dominance and Suboptimality Detection
Submitted to INFORMS Journal on Computing
null
null
null
cs.DS cs.AI cs.DM math.OC
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The branch-and-bound algorithm based on decision diagrams introduced by Bergman et al. in 2016 is a framework for solving discrete optimization problems with a dynamic programming formulation. It works by compiling a series of bounded-width decision diagrams that can provide lower and upper bounds for any given subproblem. Eventually, every part of the search space will be either explored or pruned by the algorithm, thus proving optimality. This paper presents new ingredients to speed up the search by exploiting the structure of dynamic programming models. The key idea is to prevent the repeated expansion of nodes corresponding to the same dynamic programming states by querying expansion thresholds cached throughout the search. These thresholds are based on dominance relations between partial solutions previously found and on the pruning inequalities of the filtering techniques introduced by Gillard et al. in 2021. Computational experiments show that the pruning brought by this caching mechanism allows significantly reducing the number of nodes expanded by the algorithm. This results in more benchmark instances of difficult optimization problems being solved in less time while using narrower decision diagrams.
[ { "version": "v1", "created": "Tue, 22 Nov 2022 10:18:33 GMT" }, { "version": "v2", "created": "Fri, 26 May 2023 15:51:22 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 13:50:18 GMT" } ]
2023-10-06T00:00:00
[ [ "Coppé", "Vianney", "" ], [ "Gillard", "Xavier", "" ], [ "Schaus", "Pierre", "" ] ]
not_new_dataset
0.997333
2212.00431
Violetta Weger
Markus Grassl, Anna-Lena Horlemann, Violetta Weger
The Subfield Metric and its Application to Quantum Error Correction
null
null
10.1142/S021949882550063X
null
cs.IT math.IT quant-ph
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We introduce a new weight and corresponding metric over finite extension fields for asymmetric error correction. The weight distinguishes between elements from the base field and the ones outside of it, which is motivated by asymmetric quantum codes. We set up the theoretic framework for this weight and metric, including upper and lower bounds, asymptotic behavior of random codes, and we show the existence of an optimal family of codes achieving the Singleton-type upper bound.
[ { "version": "v1", "created": "Thu, 1 Dec 2022 11:02:31 GMT" } ]
2023-10-06T00:00:00
[ [ "Grassl", "Markus", "" ], [ "Horlemann", "Anna-Lena", "" ], [ "Weger", "Violetta", "" ] ]
not_new_dataset
0.997357
2212.02648
Mazda Moayeri
Mazda Moayeri, Wenxiao Wang, Sahil Singla, Soheil Feizi
Spuriosity Rankings: Sorting Data to Measure and Mitigate Biases
Accepted to NeurIPS '23 (Spotlight)
null
null
null
cs.CV cs.AI cs.HC cs.LG
http://creativecommons.org/licenses/by/4.0/
We present a simple but effective method to measure and mitigate model biases caused by reliance on spurious cues. Instead of requiring costly changes to one's data or model training, our method better utilizes the data one already has by sorting them. Specifically, we rank images within their classes based on spuriosity (the degree to which common spurious cues are present), proxied via deep neural features of an interpretable network. With spuriosity rankings, it is easy to identify minority subpopulations (i.e. low spuriosity images) and assess model bias as the gap in accuracy between high and low spuriosity images. One can even efficiently remove a model's bias at little cost to accuracy by finetuning its classification head on low spuriosity images, resulting in fairer treatment of samples regardless of spuriosity. We demonstrate our method on ImageNet, annotating $5000$ class-feature dependencies ($630$ of which we find to be spurious) and generating a dataset of $325k$ soft segmentations for these features along the way. Having computed spuriosity rankings via the identified spurious neural features, we assess biases for $89$ diverse models and find that class-wise biases are highly correlated across models. Our results suggest that model bias due to spurious feature reliance is influenced far more by what the model is trained on than how it is trained.
[ { "version": "v1", "created": "Mon, 5 Dec 2022 23:15:43 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 17:59:06 GMT" } ]
2023-10-06T00:00:00
[ [ "Moayeri", "Mazda", "" ], [ "Wang", "Wenxiao", "" ], [ "Singla", "Sahil", "" ], [ "Feizi", "Soheil", "" ] ]
not_new_dataset
0.996785
2212.06074
Chiyuan Zhang
Badih Ghazi, Pritish Kamath, Ravi Kumar, Ethan Leeman, Pasin Manurangsi, Avinash V Varadarajan, Chiyuan Zhang
Regression with Label Differential Privacy
Appeared at ICLR '23, 28 pages, 6 figures
null
null
null
cs.LG cs.CR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We study the task of training regression models with the guarantee of label differential privacy (DP). Based on a global prior distribution on label values, which could be obtained privately, we derive a label DP randomization mechanism that is optimal under a given regression loss function. We prove that the optimal mechanism takes the form of a "randomized response on bins", and propose an efficient algorithm for finding the optimal bin values. We carry out a thorough experimental evaluation on several datasets demonstrating the efficacy of our algorithm.
[ { "version": "v1", "created": "Mon, 12 Dec 2022 17:41:32 GMT" }, { "version": "v2", "created": "Tue, 29 Aug 2023 22:30:15 GMT" }, { "version": "v3", "created": "Wed, 4 Oct 2023 18:45:53 GMT" } ]
2023-10-06T00:00:00
[ [ "Ghazi", "Badih", "" ], [ "Kamath", "Pritish", "" ], [ "Kumar", "Ravi", "" ], [ "Leeman", "Ethan", "" ], [ "Manurangsi", "Pasin", "" ], [ "Varadarajan", "Avinash V", "" ], [ "Zhang", "Chiyuan", "" ] ]
not_new_dataset
0.997422
2212.06921
Dylan Sam
Dylan Sam, J. Zico Kolter
Losses over Labels: Weakly Supervised Learning via Direct Loss Construction
13 pages, 3 figures, AAAI 2023
null
null
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Owing to the prohibitive costs of generating large amounts of labeled data, programmatic weak supervision is a growing paradigm within machine learning. In this setting, users design heuristics that provide noisy labels for subsets of the data. These weak labels are combined (typically via a graphical model) to form pseudolabels, which are then used to train a downstream model. In this work, we question a foundational premise of the typical weakly supervised learning pipeline: given that the heuristic provides all ``label" information, why do we need to generate pseudolabels at all? Instead, we propose to directly transform the heuristics themselves into corresponding loss functions that penalize differences between our model and the heuristic. By constructing losses directly from the heuristics, we can incorporate more information than is used in the standard weakly supervised pipeline, such as how the heuristics make their decisions, which explicitly informs feature selection during training. We call our method Losses over Labels (LoL) as it creates losses directly from heuristics without going through the intermediate step of a label. We show that LoL improves upon existing weak supervision methods on several benchmark text and image classification tasks and further demonstrate that incorporating gradient information leads to better performance on almost every task.
[ { "version": "v1", "created": "Tue, 13 Dec 2022 22:29:14 GMT" }, { "version": "v2", "created": "Wed, 4 Oct 2023 23:32:44 GMT" } ]
2023-10-06T00:00:00
[ [ "Sam", "Dylan", "" ], [ "Kolter", "J. Zico", "" ] ]
not_new_dataset
0.997452
2212.12055
Haiyuan Li
Haiyuan Li, Amin Emami, Karcius Assis, Antonis Vafeas, Ruizhi Yang, Reza Nejabati, Shuangyi Yan, and Dimitra Simeonidou
DRL-based Energy-Efficient Baseband Function Deployments for Service-Oriented Open RAN
null
null
null
null
cs.NI cs.SY eess.SY
http://creativecommons.org/licenses/by/4.0/
Open Radio Access Network (Open RAN) has gained tremendous attention from industry and academia with decentralized baseband functions across multiple processing units located at different places. However, the ever-expanding scope of RANs, along with fluctuations in resource utilization across different locations and timeframes, necessitates the implementation of robust function management policies to minimize network energy consumption. Most recently developed strategies neglected the activation time and the required energy for the server activation process, while this process could offset the potential energy savings gained from server hibernation. Furthermore, user plane functions, which can be deployed on edge computing servers to provide low-latency services, have not been sufficiently considered. In this paper, a multi-agent deep reinforcement learning (DRL) based function deployment algorithm, coupled with a heuristic method, has been developed to minimize energy consumption while fulfilling multiple requests and adhering to latency and resource constraints. In an 8-MEC network, the DRL-based solution approaches the performance of the benchmark while offering up to 51% energy savings compared to existing approaches. In a larger network of 14-MEC, it maintains a 38% energy-saving advantage and ensures real-time response capabilities. Furthermore, this paper prototypes an Open RAN testbed to verify the feasibility of the proposed solution.
[ { "version": "v1", "created": "Thu, 22 Dec 2022 22:07:26 GMT" }, { "version": "v2", "created": "Fri, 30 Dec 2022 21:51:34 GMT" }, { "version": "v3", "created": "Tue, 6 Jun 2023 09:00:24 GMT" }, { "version": "v4", "created": "Sun, 18 Jun 2023 13:54:38 GMT" }, { "version": "v5", "created": "Wed, 4 Oct 2023 22:10:23 GMT" } ]
2023-10-06T00:00:00
[ [ "Li", "Haiyuan", "" ], [ "Emami", "Amin", "" ], [ "Assis", "Karcius", "" ], [ "Vafeas", "Antonis", "" ], [ "Yang", "Ruizhi", "" ], [ "Nejabati", "Reza", "" ], [ "Yan", "Shuangyi", "" ], [ "Simeonidou", "Dimitra", "" ] ]
not_new_dataset
0.997202
2301.04142
Fadime Bekmambetova
Fadime Bekmambetova and Piero Triverio
Conservation properties of a leapfrog finite-difference time-domain method for the Schr\"odinger equation
36 pages, 11 figures, 5 tables
null
10.1109/TMTT.2023.3308198.
null
cs.CE physics.comp-ph
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We study the probability and energy conservation properties of a leap-frog finite-difference time-domain (FDTD) method for solving the Schr\"odinger equation. We propose expressions for the total numerical probability and energy contained in a region, and for the flux of probability current and power through its boundary. We show that the proposed expressions satisfy the conservation of probability and energy under suitable conditions. We demonstrate their connection to the Courant-Friedrichs-Lewy condition for stability. We argue that these findings can be used for developing a modular framework for stability analysis in advanced algorithms based on FDTD for solving the Schr\"odinger equation.
[ { "version": "v1", "created": "Tue, 10 Jan 2023 16:38:58 GMT" }, { "version": "v2", "created": "Wed, 12 Jul 2023 03:45:26 GMT" } ]
2023-10-06T00:00:00
[ [ "Bekmambetova", "Fadime", "" ], [ "Triverio", "Piero", "" ] ]
not_new_dataset
0.997226
2301.04494
Inder Pal Singh
Indel Pal Singh, Enjie Ghorbel, Oyebade Oyedotun, Djamila Aouada
Multi-label Image Classification using Adaptive Graph Convolutional Networks: from a Single Domain to Multiple Domains
null
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
This paper proposes an adaptive graph-based approach for multi-label image classification. Graph-based methods have been largely exploited in the field of multi-label classification, given their ability to model label correlations. Specifically, their effectiveness has been proven not only when considering a single domain but also when taking into account multiple domains. However, the topology of the used graph is not optimal as it is pre-defined heuristically. In addition, consecutive Graph Convolutional Network (GCN) aggregations tend to destroy the feature similarity. To overcome these issues, an architecture for learning the graph connectivity in an end-to-end fashion is introduced. This is done by integrating an attention-based mechanism and a similarity-preserving strategy. The proposed framework is then extended to multiple domains using an adversarial training scheme. Numerous experiments are reported on well-known single-domain and multi-domain benchmarks. The results demonstrate that our approach achieves competitive results in terms of mean Average Precision (mAP) and model size as compared to the state-of-the-art. The code will be made publicly available.
[ { "version": "v1", "created": "Wed, 11 Jan 2023 14:42:47 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 09:28:57 GMT" } ]
2023-10-06T00:00:00
[ [ "Singh", "Indel Pal", "" ], [ "Ghorbel", "Enjie", "" ], [ "Oyedotun", "Oyebade", "" ], [ "Aouada", "Djamila", "" ] ]
not_new_dataset
0.997144
2301.04554
Wei Guo
Wei Guo, Benedetta Tondi, Mauro Barni
Universal Detection of Backdoor Attacks via Density-based Clustering and Centroids Analysis
null
null
null
null
cs.CV cs.AI
http://creativecommons.org/licenses/by/4.0/
We propose a Universal Defence against backdoor attacks based on Clustering and Centroids Analysis (CCA-UD). The goal of the defence is to reveal whether a Deep Neural Network model is subject to a backdoor attack by inspecting the training dataset. CCA-UD first clusters the samples of the training set by means of density-based clustering. Then, it applies a novel strategy to detect the presence of poisoned clusters. The proposed strategy is based on a general misclassification behaviour observed when the features of a representative example of the analysed cluster are added to benign samples. The capability of inducing a misclassification error is a general characteristic of poisoned samples, hence the proposed defence is attack-agnostic. This marks a significant difference with respect to existing defences, that, either can defend against only some types of backdoor attacks, or are effective only when some conditions on the poisoning ratio or the kind of triggering signal used by the attacker are satisfied. Experiments carried out on several classification tasks and network architectures, considering different types of backdoor attacks (with either clean or corrupted labels), and triggering signals, including both global and local triggering signals, as well as sample-specific and source-specific triggers, reveal that the proposed method is very effective to defend against backdoor attacks in all the cases, always outperforming the state of the art techniques.
[ { "version": "v1", "created": "Wed, 11 Jan 2023 16:31:38 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 13:26:33 GMT" } ]
2023-10-06T00:00:00
[ [ "Guo", "Wei", "" ], [ "Tondi", "Benedetta", "" ], [ "Barni", "Mauro", "" ] ]
not_new_dataset
0.99736
2301.05603
Shiye Lei
Shiye Lei and Dacheng Tao
A Comprehensive Survey of Dataset Distillation
Accepted by IEEE TPAMI
null
null
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Deep learning technology has developed unprecedentedly in the last decade and has become the primary choice in many application domains. This progress is mainly attributed to a systematic collaboration in which rapidly growing computing resources encourage advanced algorithms to deal with massive data. However, it has gradually become challenging to handle the unlimited growth of data with limited computing power. To this end, diverse approaches are proposed to improve data processing efficiency. Dataset distillation, a dataset reduction method, addresses this problem by synthesizing a small typical dataset from substantial data and has attracted much attention from the deep learning community. Existing dataset distillation methods can be taxonomized into meta-learning and data matching frameworks according to whether they explicitly mimic the performance of target data. Although dataset distillation has shown surprising performance in compressing datasets, there are still several limitations such as distilling high-resolution data or data with complex label spaces. This paper provides a holistic understanding of dataset distillation from multiple aspects, including distillation frameworks and algorithms, factorized dataset distillation, performance comparison, and applications. Finally, we discuss challenges and promising directions to further promote future studies on dataset distillation.
[ { "version": "v1", "created": "Fri, 13 Jan 2023 15:11:38 GMT" }, { "version": "v2", "created": "Tue, 7 Feb 2023 09:21:44 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 01:09:29 GMT" } ]
2023-10-06T00:00:00
[ [ "Lei", "Shiye", "" ], [ "Tao", "Dacheng", "" ] ]
not_new_dataset
0.997412
2301.06421
Pei-Yu Chen
Pei-Yu Chen, Myrthe L. Tielman, Dirk K.J. Heylen, Catholijn M. Jonker, M. Birna van Riemsdijk
AI Alignment Dialogues: An Interactive Approach to AI Alignment in Support Agents
Withdraw because the content of the paper has been largely revised. The newest version is very different than the submitted one
null
null
null
cs.AI cs.HC
http://creativecommons.org/licenses/by/4.0/
AI alignment is about ensuring AI systems only pursue goals and activities that are beneficial to humans. Most of the current approach to AI alignment is to learn what humans value from their behavioural data. This paper proposes a different way of looking at the notion of alignment, namely by introducing AI Alignment Dialogues: dialogues with which users and agents try to achieve and maintain alignment via interaction. We argue that alignment dialogues have a number of advantages in comparison to data-driven approaches, especially for behaviour support agents, which aim to support users in achieving their desired future behaviours rather than their current behaviours. The advantages of alignment dialogues include allowing the users to directly convey higher-level concepts to the agent, and making the agent more transparent and trustworthy. In this paper we outline the concept and high-level structure of alignment dialogues. Moreover, we conducted a qualitative focus group user study from which we developed a model that describes how alignment dialogues affect users, and created design suggestions for AI alignment dialogues. Through this we establish foundations for AI alignment dialogues and shed light on what requires further development and research.
[ { "version": "v1", "created": "Mon, 16 Jan 2023 13:19:53 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 11:15:23 GMT" } ]
2023-10-06T00:00:00
[ [ "Chen", "Pei-Yu", "" ], [ "Tielman", "Myrthe L.", "" ], [ "Heylen", "Dirk K. J.", "" ], [ "Jonker", "Catholijn M.", "" ], [ "van Riemsdijk", "M. Birna", "" ] ]
not_new_dataset
0.99743
2301.07305
Mohammed Shafae
Md Habibor Rahman (1), Erfan Yazdandoost Hamedani (1), Young-Jun Son (2), Mohammed Shafae (1) ((1) The University of Arizona, (2) Purdue University)
Graph-Theoretic Approach for Manufacturing Cybersecurity Risk Modeling and Assessment
25 pages, 10 figures
null
null
null
cs.CR cs.SY eess.SY math.OC
http://creativecommons.org/licenses/by-nc-nd/4.0/
Identifying, analyzing, and evaluating cybersecurity risks are essential to assess the vulnerabilities of modern manufacturing infrastructures and to devise effective decision-making strategies to secure critical manufacturing against potential cyberattacks. In response, this work proposes a graph-theoretic approach for risk modeling and assessment to address the lack of quantitative cybersecurity risk assessment frameworks for smart manufacturing systems. In doing so, first, threat attributes are represented using an attack graphical model derived from manufacturing cyberattack taxonomies. Attack taxonomies offer consistent structures to categorize threat attributes, and the graphical approach helps model their interdependence. Second, the graphs are analyzed to explore how threat events can propagate through the manufacturing value chain and identify the manufacturing assets that threat actors can access and compromise during a threat event. Third, the proposed method identifies the attack path that maximizes the likelihood of success and minimizes the attack detection probability, and then computes the associated cybersecurity risk. Finally, the proposed risk modeling and assessment framework is demonstrated via an interconnected smart manufacturing system illustrative example. Using the proposed approach, practitioners can identify critical connections and manufacturing assets requiring prioritized security controls and develop and deploy appropriate defense measures accordingly.
[ { "version": "v1", "created": "Wed, 18 Jan 2023 04:54:00 GMT" }, { "version": "v2", "created": "Wed, 4 Oct 2023 22:42:06 GMT" } ]
2023-10-06T00:00:00
[ [ "Rahman", "Md Habibor", "" ], [ "Hamedani", "Erfan Yazdandoost", "" ], [ "Son", "Young-Jun", "" ], [ "Shafae", "Mohammed", "" ] ]
not_new_dataset
0.997348
2301.09350
Anastasios Nentidis
Anastasios Nentidis, Thomas Chatzopoulos, Anastasia Krithara, Grigorios Tsoumakas, Georgios Paliouras
Large-scale investigation of weakly-supervised deep learning for the fine-grained semantic indexing of biomedical literature
26 pages, 5 figures, 4 tables. A more concise version
Journal of Biomedical Informatics, Volume 146, 2023, 104499, ISSN 1532-0464
10.1016/j.jbi.2023.104499
null
cs.CL cs.DL cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Objective: Semantic indexing of biomedical literature is usually done at the level of MeSH descriptors with several related but distinct biomedical concepts often grouped together and treated as a single topic. This study proposes a new method for the automated refinement of subject annotations at the level of MeSH concepts. Methods: Lacking labelled data, we rely on weak supervision based on concept occurrence in the abstract of an article, which is also enhanced by dictionary-based heuristics. In addition, we investigate deep learning approaches, making design choices to tackle the particular challenges of this task. The new method is evaluated on a large-scale retrospective scenario, based on concepts that have been promoted to descriptors. Results: In our experiments concept occurrence was the strongest heuristic achieving a macro-F1 score of about 0.63 across several labels. The proposed method improved it further by more than 4pp. Conclusion: The results suggest that concept occurrence is a strong heuristic for refining the coarse-grained labels at the level of MeSH concepts and the proposed method improves it further.
[ { "version": "v1", "created": "Mon, 23 Jan 2023 10:33:22 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 14:17:39 GMT" } ]
2023-10-06T00:00:00
[ [ "Nentidis", "Anastasios", "" ], [ "Chatzopoulos", "Thomas", "" ], [ "Krithara", "Anastasia", "" ], [ "Tsoumakas", "Grigorios", "" ], [ "Paliouras", "Georgios", "" ] ]
not_new_dataset
0.997367
2302.00589
Ghazal Kalhor
Ghazal Kalhor, Tanin Zeraati, Behnam Bahrak
Diversity dilemmas: uncovering gender and nationality biases in graduate admissions across top North American computer science programs
null
null
10.1140/epjds/s13688-023-00422-5
null
cs.CY cs.SI
http://creativecommons.org/licenses/by/4.0/
Although different organizations have defined policies towards diversity in academia, many argue that minorities are still disadvantaged in university admissions due to biases. Extensive research has been conducted on detecting partiality patterns in the academic community. However, in the last few decades, limited research has focused on assessing gender and nationality biases in graduate admission results of universities. In this study, we collected a novel and comprehensive dataset containing information on approximately 14,000 graduate students majoring in computer science (CS) at the top 25 North American universities. We used statistical hypothesis tests to determine whether there is a preference for students' gender and nationality in the admission processes. In addition to partiality patterns, we discuss the relationship between gender/nationality diversity and the scientific achievements of research teams. Consistent with previous studies, our findings show that there is no gender bias in the admission of graduate students to research groups, but we observed bias based on students' nationality.
[ { "version": "v1", "created": "Wed, 1 Feb 2023 17:02:08 GMT" }, { "version": "v2", "created": "Tue, 29 Aug 2023 19:30:27 GMT" } ]
2023-10-06T00:00:00
[ [ "Kalhor", "Ghazal", "" ], [ "Zeraati", "Tanin", "" ], [ "Bahrak", "Behnam", "" ] ]
new_dataset
0.996857
2302.00942
Han Lin
Krzysztof Choromanski, Arijit Sehanobish, Han Lin, Yunfan Zhao, Eli Berger, Tetiana Parshakova, Alvin Pan, David Watkins, Tianyi Zhang, Valerii Likhosherstov, Somnath Basu Roy Chowdhury, Avinava Dubey, Deepali Jain, Tamas Sarlos, Snigdha Chaturvedi, Adrian Weller
Efficient Graph Field Integrators Meet Point Clouds
null
ICML 2023
null
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
We present two new classes of algorithms for efficient field integration on graphs encoding point clouds. The first class, SeparatorFactorization(SF), leverages the bounded genus of point cloud mesh graphs, while the second class, RFDiffusion(RFD), uses popular epsilon-nearest-neighbor graph representations for point clouds. Both can be viewed as providing the functionality of Fast Multipole Methods (FMMs), which have had a tremendous impact on efficient integration, but for non-Euclidean spaces. We focus on geometries induced by distributions of walk lengths between points (e.g., shortest-path distance). We provide an extensive theoretical analysis of our algorithms, obtaining new results in structural graph theory as a byproduct. We also perform exhaustive empirical evaluation, including on-surface interpolation for rigid and deformable objects (particularly for mesh-dynamics modeling), Wasserstein distance computations for point clouds, and the Gromov-Wasserstein variant.
[ { "version": "v1", "created": "Thu, 2 Feb 2023 08:33:36 GMT" }, { "version": "v2", "created": "Sun, 5 Feb 2023 20:12:24 GMT" }, { "version": "v3", "created": "Wed, 12 Apr 2023 22:27:17 GMT" }, { "version": "v4", "created": "Sat, 10 Jun 2023 01:29:45 GMT" }, { "version": "v5", "created": "Wed, 21 Jun 2023 02:34:32 GMT" }, { "version": "v6", "created": "Wed, 4 Oct 2023 19:17:43 GMT" } ]
2023-10-06T00:00:00
[ [ "Choromanski", "Krzysztof", "" ], [ "Sehanobish", "Arijit", "" ], [ "Lin", "Han", "" ], [ "Zhao", "Yunfan", "" ], [ "Berger", "Eli", "" ], [ "Parshakova", "Tetiana", "" ], [ "Pan", "Alvin", "" ], [ "Watkins", "David", "" ], [ "Zhang", "Tianyi", "" ], [ "Likhosherstov", "Valerii", "" ], [ "Chowdhury", "Somnath Basu Roy", "" ], [ "Dubey", "Avinava", "" ], [ "Jain", "Deepali", "" ], [ "Sarlos", "Tamas", "" ], [ "Chaturvedi", "Snigdha", "" ], [ "Weller", "Adrian", "" ] ]
not_new_dataset
0.997327
2302.02394
Zuopeng Yang
Zuopeng Yang, Tianshu Chu, Xin Lin, Erdun Gao, Daqing Liu, Jie Yang, Chaoyue Wang
Eliminating Contextual Prior Bias for Semantic Image Editing via Dual-Cycle Diffusion
This paper has been accepted by the IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-sa/4.0/
The recent success of text-to-image generation diffusion models has also revolutionized semantic image editing, enabling the manipulation of images based on query/target texts. Despite these advancements, a significant challenge lies in the potential introduction of contextual prior bias in pre-trained models during image editing, e.g., making unexpected modifications to inappropriate regions. To address this issue, we present a novel approach called Dual-Cycle Diffusion, which generates an unbiased mask to guide image editing. The proposed model incorporates a Bias Elimination Cycle that consists of both a forward path and an inverted path, each featuring a Structural Consistency Cycle to ensure the preservation of image content during the editing process. The forward path utilizes the pre-trained model to produce the edited image, while the inverted path converts the result back to the source image. The unbiased mask is generated by comparing differences between the processed source image and the edited image to ensure that both conform to the same distribution. Our experiments demonstrate the effectiveness of the proposed method, as it significantly improves the D-CLIP score from 0.272 to 0.283. The code will be available at https://github.com/JohnDreamer/DualCycleDiffsion.
[ { "version": "v1", "created": "Sun, 5 Feb 2023 14:30:22 GMT" }, { "version": "v2", "created": "Tue, 7 Feb 2023 02:57:45 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 14:35:08 GMT" } ]
2023-10-06T00:00:00
[ [ "Yang", "Zuopeng", "" ], [ "Chu", "Tianshu", "" ], [ "Lin", "Xin", "" ], [ "Gao", "Erdun", "" ], [ "Liu", "Daqing", "" ], [ "Yang", "Jie", "" ], [ "Wang", "Chaoyue", "" ] ]
not_new_dataset
0.99726
2302.02787
Lena Mangold
Lena Mangold and Camille Roth
Generative models for two-ground-truth partitions in networks
null
null
null
null
cs.SI cond-mat.stat-mech cs.LG physics.data-an physics.soc-ph
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
A myriad of approaches have been proposed to characterise the mesoscale structure of networks - most often as a partition based on patterns variously called communities, blocks, or clusters. Clearly, distinct methods designed to detect different types of patterns may provide a variety of answers to the network's mesoscale structure. Yet, even multiple runs of a given method can sometimes yield diverse and conflicting results, producing entire landscapes of partitions which potentially include multiple (locally optimal) mesoscale explanations of the network. Such ambiguity motivates a closer look at the ability of these methods to find multiple qualitatively different 'ground truth' partitions in a network. Here, we propose the stochastic cross-block model (SCBM), a generative model which allows for two distinct partitions to be built into the mesoscale structure of a single benchmark network. We demonstrate a use case of the benchmark model by appraising the power of stochastic block models (SBMs) to detect implicitly planted coexisting bi-community and core-periphery structures of different strengths. Given our model design and experimental set-up, we find that the ability to detect the two partitions individually varies by SBM variant and that coexistence of both partitions is recovered only in a very limited number of cases. Our findings suggest that in most instances only one - in some way dominating - structure can be detected, even in the presence of other partitions. They underline the need for considering entire landscapes of partitions when different competing explanations exist and motivate future research to advance partition coexistence detection methods. Our model also contributes to the field of benchmark networks more generally by enabling further exploration of the ability of new and existing methods to detect ambiguity in the mesoscale structure of networks.
[ { "version": "v1", "created": "Mon, 6 Feb 2023 14:02:28 GMT" }, { "version": "v2", "created": "Tue, 4 Jul 2023 19:01:52 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 13:00:34 GMT" } ]
2023-10-06T00:00:00
[ [ "Mangold", "Lena", "" ], [ "Roth", "Camille", "" ] ]
not_new_dataset
0.997372
2302.02936
Alex Bie
Alex Bie, Gautam Kamath, Guojun Zhang
Private GANs, Revisited
28 pages; revisions and new experiments from TMLR camera-ready + code release at https://github.com/alexbie98/dpgan-revisit
null
null
null
cs.LG cs.CR cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We show that the canonical approach for training differentially private GANs -- updating the discriminator with differentially private stochastic gradient descent (DPSGD) -- can yield significantly improved results after modifications to training. Specifically, we propose that existing instantiations of this approach neglect to consider how adding noise only to discriminator updates inhibits discriminator training, disrupting the balance between the generator and discriminator necessary for successful GAN training. We show that a simple fix -- taking more discriminator steps between generator steps -- restores parity between the generator and discriminator and improves results. Additionally, with the goal of restoring parity, we experiment with other modifications -- namely, large batch sizes and adaptive discriminator update frequency -- to improve discriminator training and see further improvements in generation quality. Our results demonstrate that on standard image synthesis benchmarks, DPSGD outperforms all alternative GAN privatization schemes. Code: https://github.com/alexbie98/dpgan-revisit.
[ { "version": "v1", "created": "Mon, 6 Feb 2023 17:11:09 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 04:47:52 GMT" } ]
2023-10-06T00:00:00
[ [ "Bie", "Alex", "" ], [ "Kamath", "Gautam", "" ], [ "Zhang", "Guojun", "" ] ]
not_new_dataset
0.997418
2302.04054
Michael Hagmann
Michael Hagmann, Philipp Meier and Stefan Riezler
Towards Inferential Reproducibility of Machine Learning Research
Published at ICLR 2023
null
null
null
cs.LG cs.AI cs.CL stat.AP stat.ML
http://creativecommons.org/licenses/by/4.0/
Reliability of machine learning evaluation -- the consistency of observed evaluation scores across replicated model training runs -- is affected by several sources of nondeterminism which can be regarded as measurement noise. Current tendencies to remove noise in order to enforce reproducibility of research results neglect inherent nondeterminism at the implementation level and disregard crucial interaction effects between algorithmic noise factors and data properties. This limits the scope of conclusions that can be drawn from such experiments. Instead of removing noise, we propose to incorporate several sources of variance, including their interaction with data properties, into an analysis of significance and reliability of machine learning evaluation, with the aim to draw inferences beyond particular instances of trained models. We show how to use linear mixed effects models (LMEMs) to analyze performance evaluation scores, and to conduct statistical inference with a generalized likelihood ratio test (GLRT). This allows us to incorporate arbitrary sources of noise like meta-parameter variations into statistical significance testing, and to assess performance differences conditional on data properties. Furthermore, a variance component analysis (VCA) enables the analysis of the contribution of noise sources to overall variance and the computation of a reliability coefficient by the ratio of substantial to total variance.
[ { "version": "v1", "created": "Wed, 8 Feb 2023 13:47:00 GMT" }, { "version": "v2", "created": "Fri, 10 Feb 2023 10:45:09 GMT" }, { "version": "v3", "created": "Thu, 16 Feb 2023 13:56:26 GMT" }, { "version": "v4", "created": "Wed, 8 Mar 2023 11:37:27 GMT" }, { "version": "v5", "created": "Thu, 13 Apr 2023 12:10:37 GMT" }, { "version": "v6", "created": "Thu, 5 Oct 2023 14:19:32 GMT" } ]
2023-10-06T00:00:00
[ [ "Hagmann", "Michael", "" ], [ "Meier", "Philipp", "" ], [ "Riezler", "Stefan", "" ] ]
not_new_dataset
0.997504
2302.11791
Gyanendra Kumar Verma
Gyanendra K. Verma and R. K. Sharma
Additive complementary dual codes over $\mathbb{F}_{q^2}$
There has been major changes in this manuscript we will submit new one
null
null
null
cs.IT math.IT
http://creativecommons.org/licenses/by/4.0/
Shi et al. [Additive complementary dual codes over F4. Designs, Codes and Cryptography, 2022.] studied additive codes over the finite field F4 with respect to trace Hermitian and trace Euclidean inner products. In this article, we define additive codes of length n over finite field Fq2 as additive subgroups of Fn q2 where q is a prime power. We associate an additive code with a matrix called a generator matrix. We characterize trace Euclidean ACD and trace Hermitian ACD codes in terms of generator matrices over the finite field Fq2 . Also, we construct these codes over Fq2 from linear LCD codes over Fq.
[ { "version": "v1", "created": "Thu, 23 Feb 2023 06:12:14 GMT" }, { "version": "v2", "created": "Sat, 6 May 2023 17:38:14 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 09:08:46 GMT" } ]
2023-10-06T00:00:00
[ [ "Verma", "Gyanendra K.", "" ], [ "Sharma", "R. K.", "" ] ]
not_new_dataset
0.996993
2303.00047
Indranil Saha
Ratijit Mitra and Indranil Saha
Online On-Demand Multi-Robot Coverage Path Planning
null
null
null
null
cs.RO
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We present an online centralized path planning algorithm to cover a large, complex, unknown workspace with multiple homogeneous mobile robots. Our algorithm is horizon-based, synchronous, and on-demand. The recently proposed horizon-based synchronous algorithms compute all the robots' paths in each horizon, significantly increasing the computation burden in large workspaces with many robots. As a remedy, we propose an algorithm that computes the paths for a subset of robots that have traversed previously computed paths entirely (thus on-demand) and reuses the remaining paths for the other robots. We formally prove that the algorithm guarantees complete coverage of the unknown workspace. Experimental results on several standard benchmark workspaces show that our algorithm scales to hundreds of robots in large complex workspaces and consistently beats a state-of-the-art online centralized multi-robot coverage path planning algorithm in terms of the time needed to achieve complete coverage. For its validation, we perform ROS+Gazebo simulations in five 2D grid benchmark workspaces with 10 Quadcopters and 10 TurtleBots, respectively. Also, to demonstrate its practical feasibility, we conduct one indoor experiment with two real TurtleBot2 robots and one outdoor experiment with three real Quadcopters.
[ { "version": "v1", "created": "Tue, 28 Feb 2023 19:43:23 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 10:02:31 GMT" } ]
2023-10-06T00:00:00
[ [ "Mitra", "Ratijit", "" ], [ "Saha", "Indranil", "" ] ]
not_new_dataset
0.997114
2303.01338
Amira Guesmi
Amira Guesmi, Muhammad Abdullah Hanif, and Muhammad Shafique
AdvRain: Adversarial Raindrops to Attack Camera-based Smart Vision Systems
null
null
null
null
cs.CV cs.CR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Vision-based perception modules are increasingly deployed in many applications, especially autonomous vehicles and intelligent robots. These modules are being used to acquire information about the surroundings and identify obstacles. Hence, accurate detection and classification are essential to reach appropriate decisions and take appropriate and safe actions at all times. Current studies have demonstrated that "printed adversarial attacks", known as physical adversarial attacks, can successfully mislead perception models such as object detectors and image classifiers. However, most of these physical attacks are based on noticeable and eye-catching patterns for generated perturbations making them identifiable/detectable by human eye or in test drives. In this paper, we propose a camera-based inconspicuous adversarial attack (\textbf{AdvRain}) capable of fooling camera-based perception systems over all objects of the same class. Unlike mask based fake-weather attacks that require access to the underlying computing hardware or image memory, our attack is based on emulating the effects of a natural weather condition (i.e., Raindrops) that can be printed on a translucent sticker, which is externally placed over the lens of a camera. To accomplish this, we provide an iterative process based on performing a random search aiming to identify critical positions to make sure that the performed transformation is adversarial for a target classifier. Our transformation is based on blurring predefined parts of the captured image corresponding to the areas covered by the raindrop. We achieve a drop in average model accuracy of more than $45\%$ and $40\%$ on VGG19 for ImageNet and Resnet34 for Caltech-101, respectively, using only $20$ raindrops.
[ { "version": "v1", "created": "Thu, 2 Mar 2023 15:14:46 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 11:55:37 GMT" } ]
2023-10-06T00:00:00
[ [ "Guesmi", "Amira", "" ], [ "Hanif", "Muhammad Abdullah", "" ], [ "Shafique", "Muhammad", "" ] ]
not_new_dataset
0.997012
2303.02950
Ying Gao
Ying Gao, Qingqing Wu, Wen Chen, Celimuge Wu, Derrick Wing Kwan Ng, Naofal Al-Dhahir
Exploiting Intelligent Reflecting Surfaces for Interference Channels with SWIPT
30 pages, accepted by IEEE Transactions on Wireless Communications
null
10.1109/TWC.2023.3318795
null
cs.IT eess.SP math.IT
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
This paper considers intelligent reflecting surface (IRS)-aided simultaneous wireless information and power transfer (SWIPT) in a multi-user multiple-input single-output (MISO) interference channel (IFC), where multiple transmitters (Txs) serve their corresponding receivers (Rxs) in a shared spectrum with the aid of IRSs. Our goal is to maximize the sum rate of the Rxs by jointly optimizing the transmit covariance matrices at the Txs, the phase shifts at the IRSs, and the resource allocation subject to the individual energy harvesting (EH) constraints at the Rxs. Towards this goal and based on the well-known power splitting (PS) and time switching (TS) receiver structures, we consider three practical transmission schemes, namely the IRS-aided hybrid TS-PS scheme, the IRS-aided time-division multiple access (TDMA) scheme, and the IRS-aided TDMA-D scheme. The latter two schemes differ in whether the Txs employ deterministic energy signals known to all the Rxs. Despite the non-convexity of the three optimization problems corresponding to the three transmission schemes, we develop computationally efficient algorithms to address them suboptimally, respectively, by capitalizing on the techniques of alternating optimization (AO) and successive convex approximation (SCA). Moreover, we conceive feasibility checking methods for these problems, based on which the initial points for the proposed algorithms are constructed. Simulation results demonstrate that our proposed IRS-aided schemes significantly outperform their counterparts without IRSs in terms of sum rate and maximum EH requirements that can be satisfied under various setups. In addition, the IRS-aided hybrid TS-PS scheme generally achieves the best sum rate performance among the three proposed IRS-aided schemes, and if not, increasing the number of IRS elements can always accomplish it.
[ { "version": "v1", "created": "Mon, 6 Mar 2023 07:44:05 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 17:09:30 GMT" } ]
2023-10-06T00:00:00
[ [ "Gao", "Ying", "" ], [ "Wu", "Qingqing", "" ], [ "Chen", "Wen", "" ], [ "Wu", "Celimuge", "" ], [ "Ng", "Derrick Wing Kwan", "" ], [ "Al-Dhahir", "Naofal", "" ] ]
not_new_dataset
0.997425
2303.06088
Marin Scalbert
Marin Scalbert and Maria Vakalopoulou and Florent Couzini\'e-Devy
Towards domain-invariant Self-Supervised Learning with Batch Styles Standardization
Under review as conference paper
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-sa/4.0/
In Self-Supervised Learning (SSL), models are typically pretrained, fine-tuned, and evaluated on the same domains. However, they tend to perform poorly when evaluated on unseen domains, a challenge that Unsupervised Domain Generalization (UDG) seeks to address. Current UDG methods rely on domain labels, which are often challenging to collect, and domain-specific architectures that lack scalability when confronted with numerous domains, making the current methodology impractical and rigid. Inspired by contrastive-based UDG methods that mitigate spurious correlations by restricting comparisons to examples from the same domain, we hypothesize that eliminating style variability within a batch could provide a more convenient and flexible way to reduce spurious correlations without requiring domain labels. To verify this hypothesis, we introduce Batch Styles Standardization (BSS), a relatively simple yet powerful Fourier-based method to standardize the style of images in a batch specifically designed for integration with SSL methods to tackle UDG. Combining BSS with existing SSL methods offers serious advantages over prior UDG methods: (1) It eliminates the need for domain labels or domain-specific network components to enhance domain-invariance in SSL representations, and (2) offers flexibility as BSS can be seamlessly integrated with diverse contrastive-based but also non-contrastive-based SSL methods. Experiments on several UDG datasets demonstrate that it significantly improves downstream task performances on unseen domains, often outperforming or rivaling with UDG methods. Finally, this work clarifies the underlying mechanisms contributing to BSS's effectiveness in improving domain-invariance in SSL representations and performances on unseen domain.
[ { "version": "v1", "created": "Fri, 10 Mar 2023 17:09:04 GMT" }, { "version": "v2", "created": "Mon, 13 Mar 2023 10:05:01 GMT" }, { "version": "v3", "created": "Mon, 24 Apr 2023 10:04:08 GMT" }, { "version": "v4", "created": "Thu, 5 Oct 2023 09:55:46 GMT" } ]
2023-10-06T00:00:00
[ [ "Scalbert", "Marin", "" ], [ "Vakalopoulou", "Maria", "" ], [ "Couzinié-Devy", "Florent", "" ] ]
not_new_dataset
0.99744
2303.09230
Xie Yi
Yi Xie, Huaidong Zhang, Xuemiao Xu, Jianqing Zhu, Shengfeng He
Towards a Smaller Student: Capacity Dynamic Distillation for Efficient Image Retrieval
Accepted by CVPR2023
Towards a Smaller Student: Capacity Dynamic Distillation for Efficient Image Retrieval, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023,16006-16015
10.1109/CVPR52729.2023.01536
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Previous Knowledge Distillation based efficient image retrieval methods employs a lightweight network as the student model for fast inference. However, the lightweight student model lacks adequate representation capacity for effective knowledge imitation during the most critical early training period, causing final performance degeneration. To tackle this issue, we propose a Capacity Dynamic Distillation framework, which constructs a student model with editable representation capacity. Specifically, the employed student model is initially a heavy model to fruitfully learn distilled knowledge in the early training epochs, and the student model is gradually compressed during the training. To dynamically adjust the model capacity, our dynamic framework inserts a learnable convolutional layer within each residual block in the student model as the channel importance indicator. The indicator is optimized simultaneously by the image retrieval loss and the compression loss, and a retrieval-guided gradient resetting mechanism is proposed to release the gradient conflict. Extensive experiments show that our method has superior inference speed and accuracy, e.g., on the VeRi-776 dataset, given the ResNet101 as a teacher, our method saves 67.13% model parameters and 65.67% FLOPs (around 24.13% and 21.94% higher than state-of-the-arts) without sacrificing accuracy (around 2.11% mAP higher than state-of-the-arts).
[ { "version": "v1", "created": "Thu, 16 Mar 2023 11:09:22 GMT" }, { "version": "v2", "created": "Wed, 31 May 2023 15:32:48 GMT" } ]
2023-10-06T00:00:00
[ [ "Xie", "Yi", "" ], [ "Zhang", "Huaidong", "" ], [ "Xu", "Xuemiao", "" ], [ "Zhu", "Jianqing", "" ], [ "He", "Shengfeng", "" ] ]
not_new_dataset
0.997303
2303.09234
Yining Jiao
Yining Jiao, Carlton Zdanski, Julia Kimbell, Andrew Prince, Cameron Worden, Samuel Kirse, Christopher Rutter, Benjamin Shields, William Dunn, Jisan Mahmud, Marc Niethammer
NAISR: A 3D Neural Additive Model for Interpretable Shape Representation
28 pages
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-nd/4.0/
Deep implicit functions (DIFs) have emerged as a powerful paradigm for many computer vision tasks such as 3D shape reconstruction, generation, registration, completion, editing, and understanding. However, given a set of 3D shapes with associated covariates there is at present no shape representation method which allows to precisely represent the shapes while capturing the individual dependencies on each covariate. Such a method would be of high utility to researchers to discover knowledge hidden in a population of shapes. For scientific shape discovery, we propose a 3D Neural Additive Model for Interpretable Shape Representation ($\texttt{NAISR}$) which describes individual shapes by deforming a shape atlas in accordance to the effect of disentangled covariates. Our approach captures shape population trends and allows for patient-specific predictions through shape transfer. $\texttt{NAISR}$ is the first approach to combine the benefits of deep implicit shape representations with an atlas deforming according to specified covariates. We evaluate $\texttt{NAISR}$ with respect to shape reconstruction, shape disentanglement, shape evolution, and shape transfer on three datasets: 1) $\textit{Starman}$, a simulated 2D shape dataset; 2) the ADNI hippocampus 3D shape dataset; and 3) a pediatric airway 3D shape dataset. Our experiments demonstrate that $\textit{Starman}$ achieves excellent shape reconstruction performance while retaining interpretability. Our code is available at $\href{https://github.com/uncbiag/NAISR}{https://github.com/uncbiag/NAISR}$.
[ { "version": "v1", "created": "Thu, 16 Mar 2023 11:18:04 GMT" }, { "version": "v2", "created": "Sat, 18 Mar 2023 12:13:19 GMT" }, { "version": "v3", "created": "Tue, 28 Mar 2023 20:07:21 GMT" }, { "version": "v4", "created": "Thu, 5 Oct 2023 09:25:26 GMT" } ]
2023-10-06T00:00:00
[ [ "Jiao", "Yining", "" ], [ "Zdanski", "Carlton", "" ], [ "Kimbell", "Julia", "" ], [ "Prince", "Andrew", "" ], [ "Worden", "Cameron", "" ], [ "Kirse", "Samuel", "" ], [ "Rutter", "Christopher", "" ], [ "Shields", "Benjamin", "" ], [ "Dunn", "William", "" ], [ "Mahmud", "Jisan", "" ], [ "Niethammer", "Marc", "" ] ]
not_new_dataset
0.996392
2303.09874
Alexander Hepburn
Alexander Hepburn, Valero Laparra, Ra\'ul Santos-Rodriguez, Jes\'us Malo
Disentangling the Link Between Image Statistics and Human Perception
null
null
null
null
cs.CV cs.LG q-bio.NC
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In the 1950s, Barlow and Attneave hypothesised a link between biological vision and information maximisation. Following Shannon, information was defined using the probability of natural images. A number of physiological and psychophysical phenomena have been derived ever since from principles like info-max, efficient coding, or optimal denoising. However, it remains unclear how this link is expressed in mathematical terms from image probability. First, classical derivations were subjected to strong assumptions on the probability models and on the behaviour of the sensors. Moreover, the direct evaluation of the hypothesis was limited by the inability of the classical image models to deliver accurate estimates of the probability. In this work we directly evaluate image probabilities using an advanced generative model for natural images, and we analyse how probability-related factors can be combined to predict human perception via sensitivity of state-of-the-art subjective image quality metrics. We use information theory and regression analysis to find a combination of just two probability-related factors that achieves 0.8 correlation with subjective metrics. This probability-based sensitivity is psychophysically validated by reproducing the basic trends of the Contrast Sensitivity Function, its suprathreshold variation, and trends of the Weber-law and masking.
[ { "version": "v1", "created": "Fri, 17 Mar 2023 10:38:27 GMT" }, { "version": "v2", "created": "Mon, 2 Oct 2023 09:40:54 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 14:06:32 GMT" } ]
2023-10-06T00:00:00
[ [ "Hepburn", "Alexander", "" ], [ "Laparra", "Valero", "" ], [ "Santos-Rodriguez", "Raúl", "" ], [ "Malo", "Jesús", "" ] ]
not_new_dataset
0.997416
2303.10650
Natalia \'Slusarz
Natalia \'Slusarz, Ekaterina Komendantskaya, Matthew L. Daggitt, Robert Stewart, Kathrin Stark
Logic of Differentiable Logics: Towards a Uniform Semantics of DL
LPAR'23
null
null
null
cs.LO cs.AI cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Differentiable logics (DL) have recently been proposed as a method of training neural networks to satisfy logical specifications. A DL consists of a syntax in which specifications are stated and an interpretation function that translates expressions in the syntax into loss functions. These loss functions can then be used during training with standard gradient descent algorithms. The variety of existing DLs and the differing levels of formality with which they are treated makes a systematic comparative study of their properties and implementations difficult. This paper remedies this problem by suggesting a meta-language for defining DLs that we call the Logic of Differentiable Logics, or LDL. Syntactically, it generalises the syntax of existing DLs to FOL, and for the first time introduces the formalism for reasoning about vectors and learners. Semantically, it introduces a general interpretation function that can be instantiated to define loss functions arising from different existing DLs. We use LDL to establish several theoretical properties of existing DLs, and to conduct their empirical study in neural network verification.
[ { "version": "v1", "created": "Sun, 19 Mar 2023 13:03:51 GMT" }, { "version": "v2", "created": "Mon, 15 May 2023 13:30:45 GMT" }, { "version": "v3", "created": "Wed, 24 May 2023 13:33:37 GMT" }, { "version": "v4", "created": "Thu, 5 Oct 2023 11:17:08 GMT" } ]
2023-10-06T00:00:00
[ [ "Ślusarz", "Natalia", "" ], [ "Komendantskaya", "Ekaterina", "" ], [ "Daggitt", "Matthew L.", "" ], [ "Stewart", "Robert", "" ], [ "Stark", "Kathrin", "" ] ]
not_new_dataset
0.997339
2303.12214
Jingwei Zhang
Jingwei Zhang, Saarthak Kapse, Ke Ma, Prateek Prasanna, Joel Saltz, Maria Vakalopoulou, Dimitris Samaras
Prompt-MIL: Boosting Multi-Instance Learning Schemes via Task-specific Prompt Tuning
Accepted to MICCAI 2023 (Oral)
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Whole slide image (WSI) classification is a critical task in computational pathology, requiring the processing of gigapixel-sized images, which is challenging for current deep-learning methods. Current state of the art methods are based on multi-instance learning schemes (MIL), which usually rely on pretrained features to represent the instances. Due to the lack of task-specific annotated data, these features are either obtained from well-established backbones on natural images, or, more recently from self-supervised models pretrained on histopathology. However, both approaches yield task-agnostic features, resulting in performance loss compared to the appropriate task-related supervision, if available. In this paper, we show that when task-specific annotations are limited, we can inject such supervision into downstream task training, to reduce the gap between fully task-tuned and task agnostic features. We propose Prompt-MIL, an MIL framework that integrates prompts into WSI classification. Prompt-MIL adopts a prompt tuning mechanism, where only a small fraction of parameters calibrates the pretrained features to encode task-specific information, rather than the conventional full fine-tuning approaches. Extensive experiments on three WSI datasets, TCGA-BRCA, TCGA-CRC, and BRIGHT, demonstrate the superiority of Prompt-MIL over conventional MIL methods, achieving a relative improvement of 1.49%-4.03% in accuracy and 0.25%-8.97% in AUROC while using fewer than 0.3% additional parameters. Compared to conventional full fine-tuning approaches, we fine-tune less than 1.3% of the parameters, yet achieve a relative improvement of 1.29%-13.61% in accuracy and 3.22%-27.18% in AUROC and reduce GPU memory consumption by 38%-45% while training 21%-27% faster. Our code is available at https://github.com/cvlab-stonybrook/PromptMIL.
[ { "version": "v1", "created": "Tue, 21 Mar 2023 22:24:27 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 03:50:19 GMT" } ]
2023-10-06T00:00:00
[ [ "Zhang", "Jingwei", "" ], [ "Kapse", "Saarthak", "" ], [ "Ma", "Ke", "" ], [ "Prasanna", "Prateek", "" ], [ "Saltz", "Joel", "" ], [ "Vakalopoulou", "Maria", "" ], [ "Samaras", "Dimitris", "" ] ]
not_new_dataset
0.99728
2303.14655
Ji Qi
Ji Qi, Jifan Yu, Teng Tu, Kunyu Gao, Yifan Xu, Xinyu Guan, Xiaozhi Wang, Yuxiao Dong, Bin Xu, Lei Hou, Juanzi Li, Jie Tang, Weidong Guo, Hui Liu, Yu Xu
GOAL: A Challenging Knowledge-grounded Video Captioning Benchmark for Real-time Soccer Commentary Generation
Accepted by CIKM 2023
null
null
null
cs.CV cs.CL cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Despite the recent emergence of video captioning models, how to generate vivid, fine-grained video descriptions based on the background knowledge (i.e., long and informative commentary about the domain-specific scenes with appropriate reasoning) is still far from being solved, which however has great applications such as automatic sports narrative. In this paper, we present GOAL, a benchmark of over 8.9k soccer video clips, 22k sentences, and 42k knowledge triples for proposing a challenging new task setting as Knowledge-grounded Video Captioning (KGVC). Moreover, we conduct experimental adaption of existing methods to show the difficulty and potential directions for solving this valuable and applicable task. Our data and code are available at https://github.com/THU-KEG/goal.
[ { "version": "v1", "created": "Sun, 26 Mar 2023 08:43:36 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 06:55:13 GMT" } ]
2023-10-06T00:00:00
[ [ "Qi", "Ji", "" ], [ "Yu", "Jifan", "" ], [ "Tu", "Teng", "" ], [ "Gao", "Kunyu", "" ], [ "Xu", "Yifan", "" ], [ "Guan", "Xinyu", "" ], [ "Wang", "Xiaozhi", "" ], [ "Dong", "Yuxiao", "" ], [ "Xu", "Bin", "" ], [ "Hou", "Lei", "" ], [ "Li", "Juanzi", "" ], [ "Tang", "Jie", "" ], [ "Guo", "Weidong", "" ], [ "Liu", "Hui", "" ], [ "Xu", "Yu", "" ] ]
new_dataset
0.997324
2303.15375
Yan Sun
Yan Sun, Yifan Yuan, Zeduo Yu, Reese Kuper, Chihun Song, Jinghan Huang, Houxiang Ji, Siddharth Agarwal, Jiaqi Lou, Ipoom Jeong, Ren Wang, Jung Ho Ahn, Tianyin Xu, Nam Sung Kim
Demystifying CXL Memory with Genuine CXL-Ready Systems and Devices
This paper has been accepted by MICRO'23. Please refer to the https://doi.org/10.1145/3613424.3614256 for the official version of this paper
null
10.1145/3613424.3614256
null
cs.PF cs.AR
http://creativecommons.org/licenses/by/4.0/
The ever-growing demands for memory with larger capacity and higher bandwidth have driven recent innovations on memory expansion and disaggregation technologies based on Compute eXpress Link (CXL). Especially, CXL-based memory expansion technology has recently gained notable attention for its ability not only to economically expand memory capacity and bandwidth but also to decouple memory technologies from a specific memory interface of the CPU. However, since CXL memory devices have not been widely available, they have been emulated using DDR memory in a remote NUMA node. In this paper, for the first time, we comprehensively evaluate a true CXL-ready system based on the latest 4th-generation Intel Xeon CPU with three CXL memory devices from different manufacturers. Specifically, we run a set of microbenchmarks not only to compare the performance of true CXL memory with that of emulated CXL memory but also to analyze the complex interplay between the CPU and CXL memory in depth. This reveals important differences between emulated CXL memory and true CXL memory, some of which will compel researchers to revisit the analyses and proposals from recent work. Next, we identify opportunities for memory-bandwidth-intensive applications to benefit from the use of CXL memory. Lastly, we propose a CXL-memory-aware dynamic page allocation policy, Caption to more efficiently use CXL memory as a bandwidth expander. We demonstrate that Caption can automatically converge to an empirically favorable percentage of pages allocated to CXL memory, which improves the performance of memory-bandwidth-intensive applications by up to 24% when compared to the default page allocation policy designed for traditional NUMA systems.
[ { "version": "v1", "created": "Mon, 27 Mar 2023 16:51:26 GMT" }, { "version": "v2", "created": "Tue, 4 Apr 2023 04:25:32 GMT" }, { "version": "v3", "created": "Sun, 30 Jul 2023 22:40:13 GMT" }, { "version": "v4", "created": "Thu, 5 Oct 2023 03:58:56 GMT" } ]
2023-10-06T00:00:00
[ [ "Sun", "Yan", "" ], [ "Yuan", "Yifan", "" ], [ "Yu", "Zeduo", "" ], [ "Kuper", "Reese", "" ], [ "Song", "Chihun", "" ], [ "Huang", "Jinghan", "" ], [ "Ji", "Houxiang", "" ], [ "Agarwal", "Siddharth", "" ], [ "Lou", "Jiaqi", "" ], [ "Jeong", "Ipoom", "" ], [ "Wang", "Ren", "" ], [ "Ahn", "Jung Ho", "" ], [ "Xu", "Tianyin", "" ], [ "Kim", "Nam Sung", "" ] ]
not_new_dataset
0.997382
2303.16887
Guan Zhe Hong
Guan Zhe Hong, Yin Cui, Ariel Fuxman, Stanley H. Chan, Enming Luo
Towards Understanding the Effect of Pretraining Label Granularity
null
null
null
null
cs.CV cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In this paper, we study how the granularity of pretraining labels affects the generalization of deep neural networks in image classification tasks. We focus on the "fine-to-coarse" transfer learning setting, where the pretraining label space is more fine-grained than that of the target problem. Empirically, we show that pretraining on the leaf labels of ImageNet21k produces better transfer results on ImageNet1k than pretraining on other coarser granularity levels, which supports the common practice used in the community. Theoretically, we explain the benefit of fine-grained pretraining by proving that, for a data distribution satisfying certain hierarchy conditions, 1) coarse-grained pretraining only allows a neural network to learn the "common" or "easy-to-learn" features well, while 2) fine-grained pretraining helps the network learn the "rarer" or "fine-grained" features in addition to the common ones, thus improving its accuracy on hard downstream test samples in which common features are missing or weak in strength. Furthermore, we perform comprehensive experiments using the label hierarchies of iNaturalist 2021 and observe that the following conditions, in addition to proper choice of label granularity, enable the transfer to work well in practice: 1) the pretraining dataset needs to have a meaningful label hierarchy, and 2) the pretraining and target label functions need to align well.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 17:56:36 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 17:32:26 GMT" } ]
2023-10-06T00:00:00
[ [ "Hong", "Guan Zhe", "" ], [ "Cui", "Yin", "" ], [ "Fuxman", "Ariel", "" ], [ "Chan", "Stanley H.", "" ], [ "Luo", "Enming", "" ] ]
not_new_dataset
0.997515
2304.03752
Jiaqi Wang
Jiaqi Wang, Pan Zhang, Tao Chu, Yuhang Cao, Yujie Zhou, Tong Wu, Bin Wang, Conghui He, Dahua Lin
V3Det: Vast Vocabulary Visual Detection Dataset
ICCV 2023 Oral Camera Ready
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Recent advances in detecting arbitrary objects in the real world are trained and evaluated on object detection datasets with a relatively restricted vocabulary. To facilitate the development of more general visual object detection, we propose V3Det, a vast vocabulary visual detection dataset with precisely annotated bounding boxes on massive images. V3Det has several appealing properties: 1) Vast Vocabulary: It contains bounding boxes of objects from 13,204 categories on real-world images, which is 10 times larger than the existing large vocabulary object detection dataset, e.g., LVIS. 2) Hierarchical Category Organization: The vast vocabulary of V3Det is organized by a hierarchical category tree which annotates the inclusion relationship among categories, encouraging the exploration of category relationships in vast and open vocabulary object detection. 3) Rich Annotations: V3Det comprises precisely annotated objects in 243k images and professional descriptions of each category written by human experts and a powerful chatbot. By offering a vast exploration space, V3Det enables extensive benchmarks on both vast and open vocabulary object detection, leading to new observations, practices, and insights for future research. It has the potential to serve as a cornerstone dataset for developing more general visual perception systems. V3Det is available at https://v3det.openxlab.org.cn/.
[ { "version": "v1", "created": "Fri, 7 Apr 2023 17:45:35 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 12:18:14 GMT" } ]
2023-10-06T00:00:00
[ [ "Wang", "Jiaqi", "" ], [ "Zhang", "Pan", "" ], [ "Chu", "Tao", "" ], [ "Cao", "Yuhang", "" ], [ "Zhou", "Yujie", "" ], [ "Wu", "Tong", "" ], [ "Wang", "Bin", "" ], [ "He", "Conghui", "" ], [ "Lin", "Dahua", "" ] ]
new_dataset
0.997916
2304.04327
Jinyi Ye
Jinyi Ye, Nikhil Jindal, Francesco Pierri, Luca Luceri
Online Networks of Support in Distressed Environments: Solidarity and Mobilization during the Russian Invasion of Ukraine
Presented at ICWSM2023 Workshop "Data for the Wellbeing of Most Vulnerable"
Proceedings of the ICWSM Workshops 2023
10.36190/2023.05
null
cs.SI
http://creativecommons.org/licenses/by-nc-sa/4.0/
Despite their drawbacks and unintended consequences, social media networks have recently emerged as a crucial resource for individuals in distress, particularly during times of crisis. These platforms serve as a means to seek assistance and support, share reliable information, and appeal for action and solidarity. In this paper, we examine the online networks of support during the Russia-Ukraine conflict by analyzing four major social media networks: Twitter, Facebook, Instagram, and YouTube. Using a large dataset of 68 million posts, we explore the temporal patterns and interconnectedness between these platforms and online support websites. Our analysis highlights the prevalence of crowdsourcing and crowdfunding websites as the two main support platforms to mobilize resources and solicit donations, revealing their purpose and contents, and investigating different support-seeking and -receiving practices. Overall, our study underscores the potential of social media in facilitating online support in distressed environments through grassroots mobilization, contributing to the growing body of research on the positive impact of online platforms in promoting social good and protecting vulnerable populations during times of crisis and conflict.
[ { "version": "v1", "created": "Sun, 9 Apr 2023 23:27:59 GMT" }, { "version": "v2", "created": "Mon, 15 May 2023 22:17:40 GMT" }, { "version": "v3", "created": "Wed, 4 Oct 2023 21:59:32 GMT" } ]
2023-10-06T00:00:00
[ [ "Ye", "Jinyi", "" ], [ "Jindal", "Nikhil", "" ], [ "Pierri", "Francesco", "" ], [ "Luceri", "Luca", "" ] ]
not_new_dataset
0.99743
2304.05128
Xinyun Chen
Xinyun Chen, Maxwell Lin, Nathanael Sch\"arli, Denny Zhou
Teaching Large Language Models to Self-Debug
null
null
null
null
cs.CL cs.AI
http://creativecommons.org/licenses/by/4.0/
Large language models (LLMs) have achieved impressive performance on code generation. However, for complex programming tasks, generating the correct solution in one go becomes challenging, thus some prior works have designed program repair approaches to improve code generation performance. In this work, we propose Self-Debugging, which teaches a large language model to debug its predicted program via few-shot demonstrations. In particular, we demonstrate that Self-Debugging can teach the large language model to perform rubber duck debugging; i.e., without any human feedback on the code correctness or error messages, the model is able to identify its mistakes by investigating the execution results and explaining the generated code in natural language. Self-Debugging achieves the state-of-the-art performance on several code generation benchmarks, including the Spider dataset for text-to-SQL generation, TransCoder for C++-to-Python translation, and MBPP for text-to-Python generation. On the Spider benchmark where there are no unit tests to verify the correctness of predictions, Self-Debugging with code explanation consistently improves the baseline by 2-3%, and improves the prediction accuracy on problems of the hardest level by 9%. On TransCoder and MBPP where unit tests are available, Self-Debugging improves the baseline accuracy by up to 12%. Meanwhile, by leveraging feedback messages and reusing failed predictions, Self-Debugging notably improves sample efficiency, and can match or outperform baseline models that generate more than 10x candidate programs.
[ { "version": "v1", "created": "Tue, 11 Apr 2023 10:43:43 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 09:12:07 GMT" } ]
2023-10-06T00:00:00
[ [ "Chen", "Xinyun", "" ], [ "Lin", "Maxwell", "" ], [ "Schärli", "Nathanael", "" ], [ "Zhou", "Denny", "" ] ]
not_new_dataset
0.977384
2304.06715
Jonathan Crabb\'e
Jonathan Crabb\'e, Mihaela van der Schaar
Evaluating the Robustness of Interpretability Methods through Explanation Invariance and Equivariance
Presented at NeurIPS 2023
null
null
null
cs.LG cs.AI cs.CG
http://creativecommons.org/licenses/by/4.0/
Interpretability methods are valuable only if their explanations faithfully describe the explained model. In this work, we consider neural networks whose predictions are invariant under a specific symmetry group. This includes popular architectures, ranging from convolutional to graph neural networks. Any explanation that faithfully explains this type of model needs to be in agreement with this invariance property. We formalize this intuition through the notion of explanation invariance and equivariance by leveraging the formalism from geometric deep learning. Through this rigorous formalism, we derive (1) two metrics to measure the robustness of any interpretability method with respect to the model symmetry group; (2) theoretical robustness guarantees for some popular interpretability methods and (3) a systematic approach to increase the invariance of any interpretability method with respect to a symmetry group. By empirically measuring our metrics for explanations of models associated with various modalities and symmetry groups, we derive a set of 5 guidelines to allow users and developers of interpretability methods to produce robust explanations.
[ { "version": "v1", "created": "Thu, 13 Apr 2023 17:59:03 GMT" }, { "version": "v2", "created": "Fri, 12 May 2023 17:59:25 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 15:29:01 GMT" } ]
2023-10-06T00:00:00
[ [ "Crabbé", "Jonathan", "" ], [ "van der Schaar", "Mihaela", "" ] ]
not_new_dataset
0.997409
2304.08247
Keno Bressem
Tianyu Han and Lisa C. Adams and Jens-Michalis Papaioannou and Paul Grundmann and Tom Oberhauser and Alexander L\"oser and Daniel Truhn and Keno K. Bressem
MedAlpaca -- An Open-Source Collection of Medical Conversational AI Models and Training Data
null
null
null
null
cs.CL cs.AI
http://creativecommons.org/licenses/by/4.0/
As large language models (LLMs) like OpenAI's GPT series continue to make strides, we witness the emergence of artificial intelligence applications in an ever-expanding range of fields. In medicine, these LLMs hold considerable promise for improving medical workflows, diagnostics, patient care, and education. Yet, there is an urgent need for open-source models that can be deployed on-premises to safeguard patient privacy. In our work, we present an innovative dataset consisting of over 160,000 entries, specifically crafted to fine-tune LLMs for effective medical applications. We investigate the impact of fine-tuning these datasets on publicly accessible pre-trained LLMs, and subsequently, we juxtapose the performance of pre-trained-only models against the fine-tuned models concerning the examinations that future medical doctors must pass to achieve certification.
[ { "version": "v1", "created": "Fri, 14 Apr 2023 11:28:08 GMT" }, { "version": "v2", "created": "Wed, 4 Oct 2023 23:28:00 GMT" } ]
2023-10-06T00:00:00
[ [ "Han", "Tianyu", "" ], [ "Adams", "Lisa C.", "" ], [ "Papaioannou", "Jens-Michalis", "" ], [ "Grundmann", "Paul", "" ], [ "Oberhauser", "Tom", "" ], [ "Löser", "Alexander", "" ], [ "Truhn", "Daniel", "" ], [ "Bressem", "Keno K.", "" ] ]
new_dataset
0.997839
2304.08979
Xinyue Shen
Xinyue Shen and Zeyuan Chen and Michael Backes and Yang Zhang
In ChatGPT We Trust? Measuring and Characterizing the Reliability of ChatGPT
null
null
null
null
cs.CR cs.LG
http://creativecommons.org/licenses/by/4.0/
The way users acquire information is undergoing a paradigm shift with the advent of ChatGPT. Unlike conventional search engines, ChatGPT retrieves knowledge from the model itself and generates answers for users. ChatGPT's impressive question-answering (QA) capability has attracted more than 100 million users within a short period of time but has also raised concerns regarding its reliability. In this paper, we perform the first large-scale measurement of ChatGPT's reliability in the generic QA scenario with a carefully curated set of 5,695 questions across ten datasets and eight domains. We find that ChatGPT's reliability varies across different domains, especially underperforming in law and science questions. We also demonstrate that system roles, originally designed by OpenAI to allow users to steer ChatGPT's behavior, can impact ChatGPT's reliability in an imperceptible way. We further show that ChatGPT is vulnerable to adversarial examples, and even a single character change can negatively affect its reliability in certain cases. We believe that our study provides valuable insights into ChatGPT's reliability and underscores the need for strengthening the reliability and security of large language models (LLMs).
[ { "version": "v1", "created": "Tue, 18 Apr 2023 13:20:45 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 13:27:12 GMT" } ]
2023-10-06T00:00:00
[ [ "Shen", "Xinyue", "" ], [ "Chen", "Zeyuan", "" ], [ "Backes", "Michael", "" ], [ "Zhang", "Yang", "" ] ]
not_new_dataset
0.997463
2304.09666
Marc Fuchs
Marc Fuchs and Fabian Kuhn
List Defective Colorings: Distributed Algorithms and Applications
null
null
10.4230/LIPIcs.DISC.2023.22
null
cs.DC cs.DS
http://creativecommons.org/licenses/by/4.0/
The distributed coloring problem is at the core of the area of distributed graph algorithms and it is a problem that has seen tremendous progress over the last few years. Much of the remarkable recent progress on deterministic distributed coloring algorithms is based on two main tools: a) defective colorings in which every node of a given color can have a limited number of neighbors of the same color and b) list coloring, a natural generalization of the standard coloring problem that naturally appears when colorings are computed in different stages and one has to extend a previously computed partial coloring to a full coloring. In this paper, we introduce 'list defective colorings', which can be seen as a generalization of these two coloring variants. Essentially, in a list defective coloring instance, each node $v$ is given a list of colors $x_{v,1},\dots,x_{v,p}$ together with a list of defects $d_{v,1},\dots,d_{v,p}$ such that if $v$ is colored with color $x_{v, i}$, it is allowed to have at most $d_{v, i}$ neighbors with color $x_{v, i}$. We highlight the important role of list defective colorings by showing that faster list defective coloring algorithms would directly lead to faster deterministic $(\Delta+1)$-coloring algorithms in the LOCAL model. Further, we extend a recent distributed list coloring algorithm by Maus and Tonoyan [DISC '20]. Slightly simplified, we show that if for each node $v$ it holds that $\sum_{i=1}^p \big(d_{v,i}+1)^2 > \mathrm{deg}_G^2(v)\cdot polylog\Delta$ then this list defective coloring instance can be solved in a communication-efficient way in only $O(\log\Delta)$ communication rounds. This leads to the first deterministic $(\Delta+1)$-coloring algorithm in the standard CONGEST model with a time complexity of $O(\sqrt{\Delta}\cdot polylog \Delta+\log^* n)$, matching the best time complexity in the LOCAL model up to a $polylog\Delta$ factor.
[ { "version": "v1", "created": "Wed, 19 Apr 2023 13:52:47 GMT" }, { "version": "v2", "created": "Mon, 7 Aug 2023 14:23:40 GMT" } ]
2023-10-06T00:00:00
[ [ "Fuchs", "Marc", "" ], [ "Kuhn", "Fabian", "" ] ]
not_new_dataset
0.997405
2304.11004
Huayu Li
Huayu Li, Xiwen Chen, Gregory Ditzler, Janet Roveda, Ao Li
Knowledge Distillation Under Ideal Joint Classifier Assumption
null
null
null
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Knowledge distillation constitutes a potent methodology for condensing substantial neural networks into more compact and efficient counterparts. Within this context, softmax regression representation learning serves as a widely embraced approach, leveraging a pre-established teacher network to guide the learning process of a diminutive student network. Notably, despite the extensive inquiry into the efficacy of softmax regression representation learning, the intricate underpinnings governing the knowledge transfer mechanism remain inadequately elucidated. This study introduces the 'Ideal Joint Classifier Knowledge Distillation' (IJCKD) framework, an overarching paradigm that not only furnishes a lucid and exhaustive comprehension of prevailing knowledge distillation techniques but also establishes a theoretical underpinning for prospective investigations. Employing mathematical methodologies derived from domain adaptation theory, this investigation conducts a comprehensive examination of the error boundary of the student network contingent upon the teacher network. Consequently, our framework facilitates efficient knowledge transference between teacher and student networks, thereby accommodating a diverse spectrum of applications.
[ { "version": "v1", "created": "Wed, 19 Apr 2023 21:06:00 GMT" }, { "version": "v2", "created": "Wed, 4 Oct 2023 23:33:35 GMT" } ]
2023-10-06T00:00:00
[ [ "Li", "Huayu", "" ], [ "Chen", "Xiwen", "" ], [ "Ditzler", "Gregory", "" ], [ "Roveda", "Janet", "" ], [ "Li", "Ao", "" ] ]
not_new_dataset
0.997422
2304.14420
Albert Lam
Albert Lam, Mihai Anitescu, Anirudh Subramanyam
Network Cascade Vulnerability using Constrained Bayesian Optimization
13 pages, 5 figures
null
null
null
cs.SI cs.LG math.OC stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Measures of power grid vulnerability are often assessed by the amount of damage an adversary can exact on the network. However, the cascading impact of such attacks is often overlooked, even though cascades are one of the primary causes of large-scale blackouts. This paper explores modifications of transmission line protection settings as candidates for adversarial attacks, which can remain undetectable as long as the network equilibrium state remains unaltered. This forms the basis of a black-box function in a Bayesian optimization procedure, where the objective is to find protection settings that maximize network degradation due to cascading. Notably, our proposed method is agnostic to the choice of the cascade simulator and its underlying assumptions. Numerical experiments reveal that, against conventional wisdom, maximally misconfiguring the protection settings of all network lines does not cause the most cascading. More surprisingly, even when the degree of misconfiguration is limited due to resource constraints, it is still possible to find settings that produce cascades comparable in severity to instances where there are no resource constraints.
[ { "version": "v1", "created": "Thu, 27 Apr 2023 02:31:20 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 02:19:18 GMT" } ]
2023-10-06T00:00:00
[ [ "Lam", "Albert", "" ], [ "Anitescu", "Mihai", "" ], [ "Subramanyam", "Anirudh", "" ] ]
not_new_dataset
0.997363
2304.14993
Dhruv Kumar
Ishika Joshi, Ritvik Budhiraja, Harshal Dev, Jahnvi Kadia, M. Osama Ataullah, Sayan Mitra, Dhruv Kumar, Harshal D. Akolekar
ChatGPT in the Classroom: An Analysis of Its Strengths and Weaknesses for Solving Undergraduate Computer Science Questions
Accepted in SIGCSE TS 2024
null
null
null
cs.HC cs.AI cs.CY
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
ChatGPT is an AI language model developed by OpenAI that can understand and generate human-like text. It can be used for a variety of use cases such as language generation, question answering, text summarization, chatbot development, language translation, sentiment analysis, content creation, personalization, text completion, and storytelling. While ChatGPT has garnered significant positive attention, it has also generated a sense of apprehension and uncertainty in academic circles. There is concern that students may leverage ChatGPT to complete take-home assignments and exams and obtain favorable grades without genuinely acquiring knowledge. This paper adopts a quantitative approach to demonstrate ChatGPT's high degree of unreliability in answering a diverse range of questions pertaining to topics in undergraduate computer science. Our analysis shows that students may risk self-sabotage by blindly depending on ChatGPT to complete assignments and exams. We build upon this analysis to provide constructive recommendations to both students and instructors.
[ { "version": "v1", "created": "Fri, 28 Apr 2023 17:26:32 GMT" }, { "version": "v2", "created": "Wed, 17 May 2023 14:44:32 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 04:18:28 GMT" } ]
2023-10-06T00:00:00
[ [ "Joshi", "Ishika", "" ], [ "Budhiraja", "Ritvik", "" ], [ "Dev", "Harshal", "" ], [ "Kadia", "Jahnvi", "" ], [ "Ataullah", "M. Osama", "" ], [ "Mitra", "Sayan", "" ], [ "Kumar", "Dhruv", "" ], [ "Akolekar", "Harshal D.", "" ] ]
not_new_dataset
0.997434
2305.06410
Hsueh-Ti Derek Liu
Hsueh-Ti Derek Liu, Mark Gillespie, Benjamin Chislett, Nicholas Sharp, Alec Jacobson, Keenan Crane
Surface Simplification using Intrinsic Error Metrics
SIGGRAPH 2023
ACM Transactions on Graphics, Vol.42, No. 4, August 2023
null
null
cs.GR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
This paper describes a method for fast simplification of surface meshes. Whereas past methods focus on visual appearance, our goal is to solve equations on the surface. Hence, rather than approximate the extrinsic geometry, we construct a coarse intrinsic triangulation of the input domain. In the spirit of the quadric error metric (QEM), we perform greedy decimation while agglomerating global information about approximation error. In lieu of extrinsic quadrics, however, we store intrinsic tangent vectors that track how far curvature "drifts" during simplification. This process also yields a bijective map between the fine and coarse mesh, and prolongation operators for both scalar- and vector-valued data. Moreover, we obtain hard guarantees on element quality via intrinsic retriangulation - a feature unique to the intrinsic setting. The overall payoff is a "black box" approach to geometry processing, which decouples mesh resolution from the size of matrices used to solve equations. We show how our method benefits several fundamental tasks, including geometric multigrid, all-pairs geodesic distance, mean curvature flow, geodesic Voronoi diagrams, and the discrete exponential map.
[ { "version": "v1", "created": "Wed, 10 May 2023 18:41:48 GMT" }, { "version": "v2", "created": "Wed, 4 Oct 2023 18:11:24 GMT" } ]
2023-10-06T00:00:00
[ [ "Liu", "Hsueh-Ti Derek", "" ], [ "Gillespie", "Mark", "" ], [ "Chislett", "Benjamin", "" ], [ "Sharp", "Nicholas", "" ], [ "Jacobson", "Alec", "" ], [ "Crane", "Keenan", "" ] ]
not_new_dataset
0.996532
2305.07962
Constantin Runge
Constantin Runge and Thomas Wiegart and Diego Lentner
Improved List Decoding for Polar-Coded Probabilistic Shaping
5 pages, 3 figures; as presented at ISTC 2023
null
null
null
cs.IT math.IT
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
A modified successive cancellation list (SCL) decoder is proposed for polar-coded probabilistic shaping. The decoder exploits the deterministic encoding rule for shaping bits to rule out candidate code words that the encoder would not generate. This provides error detection and decreases error rates compared to standard SCL decoding while at the same time reducing the length of the outer cyclic redundancy check code.
[ { "version": "v1", "created": "Sat, 13 May 2023 16:41:56 GMT" }, { "version": "v2", "created": "Tue, 22 Aug 2023 08:38:12 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 14:37:51 GMT" } ]
2023-10-06T00:00:00
[ [ "Runge", "Constantin", "" ], [ "Wiegart", "Thomas", "" ], [ "Lentner", "Diego", "" ] ]
not_new_dataset
0.996789
2305.11779
Huitong Pan
Huitong Pan, Qi Zhang, Eduard Dragut, Cornelia Caragea, Longin Jan Latecki
DMDD: A Large-Scale Dataset for Dataset Mentions Detection
Pre-MIT Press publication version. Submitted to TACL
Transactions of the Association for Computational Linguistics. 11 (2023) 1132-1146
10.1162/tacl_a_00592
null
cs.CL cs.AI
http://creativecommons.org/licenses/by/4.0/
The recognition of dataset names is a critical task for automatic information extraction in scientific literature, enabling researchers to understand and identify research opportunities. However, existing corpora for dataset mention detection are limited in size and naming diversity. In this paper, we introduce the Dataset Mentions Detection Dataset (DMDD), the largest publicly available corpus for this task. DMDD consists of the DMDD main corpus, comprising 31,219 scientific articles with over 449,000 dataset mentions weakly annotated in the format of in-text spans, and an evaluation set, which comprises of 450 scientific articles manually annotated for evaluation purposes. We use DMDD to establish baseline performance for dataset mention detection and linking. By analyzing the performance of various models on DMDD, we are able to identify open problems in dataset mention detection. We invite the community to use our dataset as a challenge to develop novel dataset mention detection models.
[ { "version": "v1", "created": "Fri, 19 May 2023 16:18:00 GMT" } ]
2023-10-06T00:00:00
[ [ "Pan", "Huitong", "" ], [ "Zhang", "Qi", "" ], [ "Dragut", "Eduard", "" ], [ "Caragea", "Cornelia", "" ], [ "Latecki", "Longin Jan", "" ] ]
new_dataset
0.997837
2305.12081
Zifeng Wang
Zifeng Wang and Chufan Gao and Cao Xiao and Jimeng Sun
MediTab: Scaling Medical Tabular Data Predictors via Data Consolidation, Enrichment, and Refinement
null
null
null
null
cs.LG cs.AI
http://creativecommons.org/licenses/by/4.0/
Tabular data prediction has been employed in medical applications such as patient health risk prediction. However, existing methods usually revolve around the algorithm design while overlooking the significance of data engineering. Medical tabular datasets frequently exhibit significant heterogeneity across different sources, with limited sample sizes per source. As such, previous predictors are often trained on manually curated small datasets that struggle to generalize across different tabular datasets during inference. This paper proposes to scale medical tabular data predictors (MediTab) to various tabular inputs with varying features. The method uses a data engine that leverages large language models (LLMs) to consolidate tabular samples to overcome the barrier across tables with distinct schema. It also aligns out-domain data with the target task using a "learn, annotate, and refinement" pipeline. The expanded training data then enables the pre-trained MediTab to infer for arbitrary tabular input in the domain without fine-tuning, resulting in significant improvements over supervised baselines: it reaches an average ranking of 1.57 and 1.00 on 7 patient outcome prediction datasets and 3 trial outcome prediction datasets, respectively. In addition, MediTab exhibits impressive zero-shot performances: it outperforms supervised XGBoost models by 8.9% and 17.2% on average in two prediction tasks, respectively. The code is available at https://github.com/RyanWangZf/MediTab.
[ { "version": "v1", "created": "Sat, 20 May 2023 03:37:09 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 05:40:00 GMT" } ]
2023-10-06T00:00:00
[ [ "Wang", "Zifeng", "" ], [ "Gao", "Chufan", "" ], [ "Xiao", "Cao", "" ], [ "Sun", "Jimeng", "" ] ]
not_new_dataset
0.99709
2305.12766
Chi Han
Chi Han, Ziqi Wang, Han Zhao, Heng Ji
Explaining Emergent In-Context Learning as Kernel Regression
9 pages, 4 figures
null
null
null
cs.CL cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
Large language models (LLMs) have initiated a paradigm shift in transfer learning. In contrast to the classic pretraining-then-finetuning procedure, in order to use LLMs for downstream prediction tasks, one only needs to provide a few demonstrations, known as in-context examples, without adding more or updating existing model parameters. This in-context learning (ICL) capability of LLMs is intriguing, and it is not yet fully understood how pretrained LLMs acquire such capabilities. In this paper, we investigate the reason why a transformer-based language model can accomplish in-context learning after pre-training on a general language corpus by proposing one hypothesis that LLMs can simulate kernel regression with internal representations when faced with in-context examples. More concretely, we first prove that Bayesian inference on in-context prompts can be asymptotically understood as kernel regression $\hat y = \sum_i y_i K(x, x_i)/\sum_i K(x, x_i)$ as the number of in-context demonstrations grows. Then, we empirically investigate the in-context behaviors of language models. We find that during ICL, the attention and hidden features in LLMs match the behaviors of a kernel regression. Finally, our theory provides insights into multiple phenomena observed in the ICL field: why retrieving demonstrative samples similar to test samples can help, why ICL performance is sensitive to the output formats, and why ICL accuracy benefits from selecting in-distribution and representative samples.
[ { "version": "v1", "created": "Mon, 22 May 2023 06:45:02 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 16:04:43 GMT" } ]
2023-10-06T00:00:00
[ [ "Han", "Chi", "" ], [ "Wang", "Ziqi", "" ], [ "Zhao", "Han", "" ], [ "Ji", "Heng", "" ] ]
not_new_dataset
0.99746
2305.13673
Zeyuan Allen-Zhu
Zeyuan Allen-Zhu, Yuanzhi Li
Physics of Language Models: Part 1, Context-Free Grammar
V2 polishes writing and adds Appendix G
null
null
null
cs.CL cs.AI cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We design controlled experiments to study HOW generative language models, like GPT, learn context-free grammars (CFGs) -- diverse language systems with a tree-like structure capturing many aspects of natural languages, programs, and logics. CFGs are as hard as pushdown automata, and can be ambiguous so that verifying if a string satisfies the rules requires dynamic programming. We construct synthetic data and demonstrate that even for difficult (long and ambiguous) CFGs, pre-trained transformers can learn to generate sentences with near-perfect accuracy and impressive diversity. More importantly, we delve into the physical principles behind how transformers learns CFGs. We discover that the hidden states within the transformer implicitly and precisely encode the CFG structure (such as putting tree node information exactly on the subtree boundary), and learn to form "boundary to boundary" attentions resembling dynamic programming. We also cover some extension of CFGs as well as the robustness aspect of transformers against grammar mistakes. Overall, our research provides a comprehensive and empirical understanding of how transformers learn CFGs, and reveals the physical mechanisms utilized by transformers to capture the structure and rules of languages.
[ { "version": "v1", "created": "Tue, 23 May 2023 04:28:16 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 01:43:23 GMT" } ]
2023-10-06T00:00:00
[ [ "Allen-Zhu", "Zeyuan", "" ], [ "Li", "Yuanzhi", "" ] ]
not_new_dataset
0.997382
2305.13716
Yuhao Liang
Yuhao Liang, Fan Yu, Yangze Li, Pengcheng Guo, Shiliang Zhang, Qian Chen, Lei Xie
BA-SOT: Boundary-Aware Serialized Output Training for Multi-Talker ASR
Accepted by INTERSPEECH 2023
null
null
null
cs.SD cs.CL eess.AS
http://creativecommons.org/licenses/by-sa/4.0/
The recently proposed serialized output training (SOT) simplifies multi-talker automatic speech recognition (ASR) by generating speaker transcriptions separated by a special token. However, frequent speaker changes can make speaker change prediction difficult. To address this, we propose boundary-aware serialized output training (BA-SOT), which explicitly incorporates boundary knowledge into the decoder via a speaker change detection task and boundary constraint loss. We also introduce a two-stage connectionist temporal classification (CTC) strategy that incorporates token-level SOT CTC to restore temporal context information. Besides typical character error rate (CER), we introduce utterance-dependent character error rate (UD-CER) to further measure the precision of speaker change prediction. Compared to original SOT, BA-SOT reduces CER/UD-CER by 5.1%/14.0%, and leveraging a pre-trained ASR model for BA-SOT model initialization further reduces CER/UD-CER by 8.4%/19.9%.
[ { "version": "v1", "created": "Tue, 23 May 2023 06:08:13 GMT" }, { "version": "v2", "created": "Tue, 30 May 2023 13:45:08 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 11:44:39 GMT" } ]
2023-10-06T00:00:00
[ [ "Liang", "Yuhao", "" ], [ "Yu", "Fan", "" ], [ "Li", "Yangze", "" ], [ "Guo", "Pengcheng", "" ], [ "Zhang", "Shiliang", "" ], [ "Chen", "Qian", "" ], [ "Xie", "Lei", "" ] ]
not_new_dataset
0.997163
2305.14979
Gabriel Kasmi
Gabriel Kasmi and Laurent Dubus and Yves-Marie Saint Drenan and Philippe Blanc
Assessment of the Reliablity of a Model's Decision by Generalizing Attribution to the Wavelet Domain
16 pages, 10 figures, 2 tables. v1 of the manuscript rejected from NeurIPS 2023, mainly due to the lack of quantitative evidence of the relevance of the proposed methodology. In the v2, we propose steps to address this issue and also plan on expanding the insertion and deletion scores for our method
null
null
null
cs.CV cs.AI stat.ML
http://creativecommons.org/licenses/by/4.0/
Neural networks have shown remarkable performance in computer vision, but their deployment in numerous scientific and technical fields is challenging due to their black-box nature. Scientists and practitioners need to evaluate the reliability of a decision, i.e., to know simultaneously if a model relies on the relevant features and whether these features are robust to image corruptions. Existing attribution methods aim to provide human-understandable explanations by highlighting important regions in the image domain, but fail to fully characterize a decision process's reliability. To bridge this gap, we introduce the Wavelet sCale Attribution Method (WCAM), a generalization of attribution from the pixel domain to the space-scale domain using wavelet transforms. Attribution in the wavelet domain reveals where {\it and} on what scales the model focuses, thus enabling us to assess whether a decision is reliable.
[ { "version": "v1", "created": "Wed, 24 May 2023 10:13:32 GMT" }, { "version": "v2", "created": "Fri, 22 Sep 2023 16:03:50 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 11:53:31 GMT" } ]
2023-10-06T00:00:00
[ [ "Kasmi", "Gabriel", "" ], [ "Dubus", "Laurent", "" ], [ "Drenan", "Yves-Marie Saint", "" ], [ "Blanc", "Philippe", "" ] ]
not_new_dataset
0.99741
2305.15070
London Lowmanstone
London Lowmanstone, Ruyuan Wan, Risako Owan, Jaehyung Kim, Dongyeop Kang
Annotation Imputation to Individualize Predictions: Initial Studies on Distribution Dynamics and Model Predictions
NLPerspectives - 2nd Workshop on Perspectivist Approaches to NLP, 39 pages, 13 figures, 13 tables
2nd Workshop on Perspectivist Approaches to NLP 2023
null
null
cs.CL
http://creativecommons.org/licenses/by-sa/4.0/
Annotating data via crowdsourcing is time-consuming and expensive. Due to these costs, dataset creators often have each annotator label only a small subset of the data. This leads to sparse datasets with examples that are marked by few annotators. The downside of this process is that if an annotator doesn't get to label a particular example, their perspective on it is missed. This is especially concerning for subjective NLP datasets where there is no single correct label: people may have different valid opinions. Thus, we propose using imputation methods to generate the opinions of all annotators for all examples, creating a dataset that does not leave out any annotator's view. We then train and prompt models, using data from the imputed dataset, to make predictions about the distribution of responses and individual annotations. In our analysis of the results, we found that the choice of imputation method significantly impacts soft label changes and distribution. While the imputation introduces noise in the prediction of the original dataset, it has shown potential in enhancing shots for prompts, particularly for low-response-rate annotators. We have made all of our code and data publicly available.
[ { "version": "v1", "created": "Wed, 24 May 2023 11:54:46 GMT" }, { "version": "v2", "created": "Fri, 8 Sep 2023 22:17:17 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 07:10:25 GMT" } ]
2023-10-06T00:00:00
[ [ "Lowmanstone", "London", "" ], [ "Wan", "Ruyuan", "" ], [ "Owan", "Risako", "" ], [ "Kim", "Jaehyung", "" ], [ "Kang", "Dongyeop", "" ] ]
not_new_dataset
0.997202
2305.15086
Jong Chul Ye
Beomsu Kim, Gihyun Kwon, Kwanyoung Kim, Jong Chul Ye
Unpaired Image-to-Image Translation via Neural Schr\"odinger Bridge
null
null
null
null
cs.CV cs.AI cs.LG stat.ML
http://creativecommons.org/licenses/by/4.0/
Diffusion models are a powerful class of generative models which simulate stochastic differential equations (SDEs) to generate data from noise. Although diffusion models have achieved remarkable progress in recent years, they have limitations in the unpaired image-to-image translation tasks due to the Gaussian prior assumption. Schr\"odinger Bridge (SB), which learns an SDE to translate between two arbitrary distributions, have risen as an attractive solution to this problem. However, none of SB models so far have been successful at unpaired translation between high-resolution images. In this work, we propose the Unpaired Neural Schr\"odinger Bridge (UNSB), which expresses SB problem as a sequence of adversarial learning problems. This allows us to incorporate advanced discriminators and regularization to learn a SB between unpaired data. We demonstrate that UNSB is scalable and successfully solves various unpaired image-to-image translation tasks. Code: \url{https://github.com/cyclomon/UNSB}
[ { "version": "v1", "created": "Wed, 24 May 2023 12:05:24 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 05:12:09 GMT" } ]
2023-10-06T00:00:00
[ [ "Kim", "Beomsu", "" ], [ "Kwon", "Gihyun", "" ], [ "Kim", "Kwanyoung", "" ], [ "Ye", "Jong Chul", "" ] ]
not_new_dataset
0.997238
2305.16102
Xinyi Wu
Xinyi Wu, Amir Ajorlou, Zihui Wu, Ali Jadbabaie
Demystifying Oversmoothing in Attention-Based Graph Neural Networks
NeurIPS 2023 spotlight. New remarks added
null
null
null
cs.LG cs.SI stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Oversmoothing in Graph Neural Networks (GNNs) refers to the phenomenon where increasing network depth leads to homogeneous node representations. While previous work has established that Graph Convolutional Networks (GCNs) exponentially lose expressive power, it remains controversial whether the graph attention mechanism can mitigate oversmoothing. In this work, we provide a definitive answer to this question through a rigorous mathematical analysis, by viewing attention-based GNNs as nonlinear time-varying dynamical systems and incorporating tools and techniques from the theory of products of inhomogeneous matrices and the joint spectral radius. We establish that, contrary to popular belief, the graph attention mechanism cannot prevent oversmoothing and loses expressive power exponentially. The proposed framework extends the existing results on oversmoothing for symmetric GCNs to a significantly broader class of GNN models, including random walk GCNs, Graph Attention Networks (GATs) and (graph) transformers. In particular, our analysis accounts for asymmetric, state-dependent and time-varying aggregation operators and a wide range of common nonlinear activation functions, such as ReLU, LeakyReLU, GELU and SiLU.
[ { "version": "v1", "created": "Thu, 25 May 2023 14:31:59 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 15:04:27 GMT" } ]
2023-10-06T00:00:00
[ [ "Wu", "Xinyi", "" ], [ "Ajorlou", "Amir", "" ], [ "Wu", "Zihui", "" ], [ "Jadbabaie", "Ali", "" ] ]
not_new_dataset
0.997448
2305.17455
Dachuan Shi
Dachuan Shi, Chaofan Tao, Anyi Rao, Zhendong Yang, Chun Yuan, Jiaqi Wang
CrossGET: Cross-Guided Ensemble of Tokens for Accelerating Vision-Language Transformers
Technical Report
null
null
null
cs.CV cs.CL
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Recent vision-language models have achieved tremendous progress far beyond what we ever expected. However, their computational costs are also dramatically growing with rapid development, especially for the large models. It makes model acceleration exceedingly critical in a scenario of limited resources. Although extensively studied for unimodal models, the acceleration for multimodal models, especially the vision-language Transformers, is relatively under-explored. To pursue more efficient and accessible vision-language Transformers, this paper introduces \textbf{Cross}-\textbf{G}uided \textbf{E}nsemble of \textbf{T}okens (\textbf{\emph{CrossGET}}), a universal acceleration framework for vision-language Transformers. This framework adaptively combines tokens through real-time, cross-modal guidance, thereby achieving substantial acceleration while keeping high performance. \textit{CrossGET} has two key innovations: 1) \textit{Cross-Guided Matching and Ensemble}. \textit{CrossGET} incorporates cross-modal guided token matching and ensemble to exploit cross-modal information effectively, only introducing cross-modal tokens with negligible extra parameters. 2) \textit{Complete-Graph Soft Matching}. In contrast to the existing bipartite soft matching approach, \textit{CrossGET} introduces a complete-graph soft matching policy to achieve more reliable token-matching results while maintaining parallelizability and high efficiency. Extensive experiments are conducted on various vision-language tasks, including image-text retrieval, visual reasoning, image captioning, and visual question answering. Performance on both classic multimodal architectures and emerging multimodal LLMs demonstrate the effectiveness and versatility of the proposed \textit{CrossGET} framework. The code will be at \url{https://github.com/sdc17/CrossGET}.
[ { "version": "v1", "created": "Sat, 27 May 2023 12:07:21 GMT" }, { "version": "v2", "created": "Wed, 4 Oct 2023 22:11:50 GMT" } ]
2023-10-06T00:00:00
[ [ "Shi", "Dachuan", "" ], [ "Tao", "Chaofan", "" ], [ "Rao", "Anyi", "" ], [ "Yang", "Zhendong", "" ], [ "Yuan", "Chun", "" ], [ "Wang", "Jiaqi", "" ] ]
not_new_dataset
0.996958
2305.20057
Lisha Chen
Lisha Chen, Heshan Fernando, Yiming Ying, Tianyi Chen
Three-Way Trade-Off in Multi-Objective Learning: Optimization, Generalization and Conflict-Avoidance
null
null
null
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
Multi-objective learning (MOL) problems often arise in emerging machine learning problems when there are multiple learning criteria, data modalities, or learning tasks. Different from single-objective learning, one of the critical challenges in MOL is the potential conflict among different objectives during the iterative optimization process. Recent works have developed various dynamic weighting algorithms for MOL such as MGDA and its variants, where the central idea is to find an update direction that avoids conflicts among objectives. Albeit its appealing intuition, empirical studies show that dynamic weighting methods may not always outperform static ones. To understand this theory-practical gap, we focus on a new stochastic variant of MGDA - the Multi-objective gradient with Double sampling (MoDo) algorithm, and study the generalization performance of the dynamic weighting-based MoDo and its interplay with optimization through the lens of algorithm stability. Perhaps surprisingly, we find that the key rationale behind MGDA -- updating along conflict-avoidant direction - may hinder dynamic weighting algorithms from achieving the optimal ${\cal O}(1/\sqrt{n})$ population risk, where $n$ is the number of training samples. We further demonstrate the impact of the variability of dynamic weights on the three-way trade-off among optimization, generalization, and conflict avoidance that is unique in MOL. We showcase the generality of our theoretical framework by analyzing other existing stochastic MOL algorithms under the framework. Experiments on various multi-task learning benchmarks are performed to demonstrate the practical applicability. Code is available at https://github.com/heshandevaka/Trade-Off-MOL.
[ { "version": "v1", "created": "Wed, 31 May 2023 17:31:56 GMT" }, { "version": "v2", "created": "Sat, 12 Aug 2023 18:29:36 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 17:41:06 GMT" } ]
2023-10-06T00:00:00
[ [ "Chen", "Lisha", "" ], [ "Fernando", "Heshan", "" ], [ "Ying", "Yiming", "" ], [ "Chen", "Tianyi", "" ] ]
not_new_dataset
0.997448
2305.20062
Matan Levy
Matan Levy, Rami Ben-Ari, Nir Darshan, Dani Lischinski
Chatting Makes Perfect: Chat-based Image Retrieval
Camera Ready version for NeurIPS 2023
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-nd/4.0/
Chats emerge as an effective user-friendly approach for information retrieval, and are successfully employed in many domains, such as customer service, healthcare, and finance. However, existing image retrieval approaches typically address the case of a single query-to-image round, and the use of chats for image retrieval has been mostly overlooked. In this work, we introduce ChatIR: a chat-based image retrieval system that engages in a conversation with the user to elicit information, in addition to an initial query, in order to clarify the user's search intent. Motivated by the capabilities of today's foundation models, we leverage Large Language Models to generate follow-up questions to an initial image description. These questions form a dialog with the user in order to retrieve the desired image from a large corpus. In this study, we explore the capabilities of such a system tested on a large dataset and reveal that engaging in a dialog yields significant gains in image retrieval. We start by building an evaluation pipeline from an existing manually generated dataset and explore different modules and training strategies for ChatIR. Our comparison includes strong baselines derived from related applications trained with Reinforcement Learning. Our system is capable of retrieving the target image from a pool of 50K images with over 78% success rate after 5 dialogue rounds, compared to 75% when questions are asked by humans, and 64% for a single shot text-to-image retrieval. Extensive evaluations reveal the strong capabilities and examine the limitations of CharIR under different settings. Project repository is available at https://github.com/levymsn/ChatIR.
[ { "version": "v1", "created": "Wed, 31 May 2023 17:38:08 GMT" }, { "version": "v2", "created": "Thu, 5 Oct 2023 16:40:02 GMT" } ]
2023-10-06T00:00:00
[ [ "Levy", "Matan", "" ], [ "Ben-Ari", "Rami", "" ], [ "Darshan", "Nir", "" ], [ "Lischinski", "Dani", "" ] ]
not_new_dataset
0.997076
2306.00709
Lakmal Meegahapola
Nathan Kammoun and Lakmal Meegahapola and Daniel Gatica-Perez
Understanding the Social Context of Eating with Multimodal Smartphone Sensing: The Role of Country Diversity
25th ACM International Conference on Multimodal Interaction (ICMI)
null
10.1145/3577190.3614129
null
cs.HC cs.CY
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Understanding the social context of eating is crucial for promoting healthy eating behaviors. Multimodal smartphone sensor data could provide valuable insights into eating behavior, particularly in mobile food diaries and mobile health apps. However, research on the social context of eating with smartphone sensor data is limited, despite extensive studies in nutrition and behavioral science. Moreover, the impact of country differences on the social context of eating, as measured by multimodal phone sensor data and self-reports, remains under-explored. To address this research gap, our study focuses on a dataset of approximately 24K self-reports on eating events provided by 678 college students in eight countries to investigate the country diversity that emerges from smartphone sensors during eating events for different social contexts (alone or with others). Our analysis revealed that while some smartphone usage features during eating events were similar across countries, others exhibited unique trends in each country. We further studied how user and country-specific factors impact social context inference by developing machine learning models with population-level (non-personalized) and hybrid (partially personalized) experimental setups. We showed that models based on the hybrid approach achieve AUC scores up to 0.75 with XGBoost models. These findings emphasize the importance of considering country differences in building and deploying machine learning models to minimize biases and improve generalization across different populations.
[ { "version": "v1", "created": "Thu, 1 Jun 2023 14:16:59 GMT" }, { "version": "v2", "created": "Sat, 12 Aug 2023 20:31:22 GMT" }, { "version": "v3", "created": "Wed, 4 Oct 2023 21:50:48 GMT" } ]
2023-10-06T00:00:00
[ [ "Kammoun", "Nathan", "" ], [ "Meegahapola", "Lakmal", "" ], [ "Gatica-Perez", "Daniel", "" ] ]
not_new_dataset
0.995858
2306.00966
Hila Chefer
Hila Chefer, Oran Lang, Mor Geva, Volodymyr Polosukhin, Assaf Shocher, Michal Irani, Inbar Mosseri, Lior Wolf
The Hidden Language of Diffusion Models
null
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Text-to-image diffusion models have demonstrated an unparalleled ability to generate high-quality, diverse images from a textual prompt. However, the internal representations learned by these models remain an enigma. In this work, we present Conceptor, a novel method to interpret the internal representation of a textual concept by a diffusion model. This interpretation is obtained by decomposing the concept into a small set of human-interpretable textual elements. Applied over the state-of-the-art Stable Diffusion model, Conceptor reveals non-trivial structures in the representations of concepts. For example, we find surprising visual connections between concepts, that transcend their textual semantics. We additionally discover concepts that rely on mixtures of exemplars, biases, renowned artistic styles, or a simultaneous fusion of multiple meanings of the concept. Through a large battery of experiments, we demonstrate Conceptor's ability to provide meaningful, robust, and faithful decompositions for a wide variety of abstract, concrete, and complex textual concepts, while allowing to naturally connect each decomposition element to its corresponding visual impact on the generated images. Our code will be available at: https://hila-chefer.github.io/Conceptor/
[ { "version": "v1", "created": "Thu, 1 Jun 2023 17:57:08 GMT" }, { "version": "v2", "created": "Tue, 6 Jun 2023 13:16:43 GMT" }, { "version": "v3", "created": "Thu, 5 Oct 2023 12:55:12 GMT" } ]
2023-10-06T00:00:00
[ [ "Chefer", "Hila", "" ], [ "Lang", "Oran", "" ], [ "Geva", "Mor", "" ], [ "Polosukhin", "Volodymyr", "" ], [ "Shocher", "Assaf", "" ], [ "Irani", "Michal", "" ], [ "Mosseri", "Inbar", "" ], [ "Wolf", "Lior", "" ] ]
not_new_dataset
0.996881