title
stringlengths
18
162
paper_url
stringlengths
42
44
authors
listlengths
1
29
type
stringclasses
3 values
primary_area
stringclasses
13 values
abstract
large_stringlengths
400
2.37k
keywords
listlengths
0
22
TL;DR
large_stringclasses
0 values
submission_number
int64
9
6.62k
arxiv_id
stringlengths
10
10
arxiv_id_source
stringclasses
3 values
embedding
listlengths
768
768
Quantifying and Mitigating the Impact of Label Errors on Model Disparity Metrics
https://openreview.net/forum?id=RUzSobdYy0V
[ "Julius Adebayo", "Melissa Hall", "Bowen Yu", "Bobbie Chern" ]
Poster
Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)
Errors in labels obtained via human annotation adversely affect a trained model's performance. Existing approaches propose ways to mitigate the effect of label error on a model's downstream accuracy, yet little is known about its impact on a model's group-based disparity metrics\footnote{Group-based disparity metrics l...
[]
null
6,620
2310.02533
title_snapshot
[ -0.008871098048985004, 0.02004285156726837, -0.03223993629217148, 0.03216385468840599, 0.03212111443281174, 0.023045269772410393, 0.02972657047212124, 0.004193576984107494, -0.036789149045944214, -0.028881344944238663, -0.024471359327435493, 0.024003850296139717, -0.08738642185926437, -0.0...
Factorized Fourier Neural Operators
https://openreview.net/forum?id=tmIiMPl4IPa
[ "Alasdair Tran", "Alexander Mathews", "Lexing Xie", "Cheng Soon Ong" ]
Poster
Machine Learning for Sciences (eg biology, physics, health sciences, social sciences, climate/sustainability )
We propose the Factorized Fourier Neural Operator (F-FNO), a learning-based approach for simulating partial differential equations (PDEs). Starting from a recently proposed Fourier representation of flow fields, the F-FNO bridges the performance gap between pure machine learning approaches to that of the best numerical...
[ "fourier transform", "fourier operators", "pde", "navier stokes" ]
null
6,610
2111.13802
title_snapshot
[ -0.047451164573431015, -0.02812906727194786, 0.03021908551454544, 0.026210660114884377, 0.028943542391061783, 0.04832187667489052, -0.015666397288441658, 0.0031966085080057383, -0.04785004258155823, -0.06070670112967491, 0.02280525676906109, -0.020632248371839523, -0.03828040137887001, 0.0...
DFPC: Data flow driven pruning of coupled channels without data.
https://openreview.net/forum?id=mhnHqRqcjYU
[ "Tanay Narshana", "Chaitanya Murti", "Chiranjib Bhattacharyya" ]
Poster
Deep Learning and representational learning
Modern, multi-branched neural network architectures often possess complex interconnections between layers, which we call coupled channels (CCs). Structured pruning of CCs in these multi-branch networks is an under-researched problem, as most existing works are typically designed for pruning single-branch models like VG...
[ "Pruning", "Data Free", "Model Compression" ]
null
6,603
null
null
[ 0.0007754036341793835, -0.01838887669146061, -0.0002316441823495552, 0.04332970082759857, 0.04419385641813278, 0.04160864278674126, -0.018240515142679214, -0.0022899960167706013, -0.01779215596616268, -0.05499108508229256, -0.005414722021669149, -0.014834927394986153, -0.05458105728030205, ...
TVSPrune - Pruning Non-discriminative filters via Total Variation separability of intermediate representations without fine tuning
https://openreview.net/forum?id=sZI1Oj9KBKy
[ "Chaitanya Murti", "Tanay Narshana", "Chiranjib Bhattacharyya" ]
Poster
Deep Learning and representational learning
Achieving structured, data-free sparsity of deep neural networks (DNNs) remains an open area of research. In this work, we address the challenge of pruning filters without access to the original training set or loss function. We propose the discriminative filters hypothesis, that well-trained models possess discrimina...
[ "Structured pruning", "model compression" ]
null
6,601
null
null
[ 0.006295611150562763, -0.017653798684477806, 0.009630952030420303, 0.026365185156464577, 0.03112618252635002, 0.03716495260596275, 0.013711030595004559, -0.017592402175068855, -0.03884253650903702, -0.05035768821835518, -0.02695184387266636, 0.020346324890851974, -0.06982611864805222, 0.02...
Finding Actual Descent Directions for Adversarial Training
https://openreview.net/forum?id=I3HCE7Ro78H
[ "Fabian Latorre", "Igor Krawczuk", "Leello Tadesse Dadi", "Thomas Pethick", "Volkan Cevher" ]
Poster
Optimization (eg, convex and non-convex optimization)
Adversarial Training using a strong first-order adversary (PGD) is the gold standard for training Deep Neural Networks that are robust to adversarial examples. We show that, contrary to the general understanding of the method, the gradient at an optimal adversarial example may increase, rather than decrease, the advers...
[ "Adversarial Training", "Adversarial Examples", "non-convex optimization", "robustness" ]
null
6,599
null
null
[ -0.04606470838189125, -0.03350912407040596, -0.008699988946318626, 0.024022947996854782, 0.02023494988679886, 0.028205621987581253, 0.042069196701049805, -0.012220893055200577, -0.020900700241327286, -0.03792285546660423, -0.004105553496629, -0.00921888742595911, -0.04305678978562355, -0.0...
Learning Continuous Normalizing Flows For Faster Convergence To Target Distribution via Ascent Regularizations
https://openreview.net/forum?id=6iEoTr-jeB7
[ "Shuangshuang Chen", "Sihao Ding", "Yiannis Karayiannidis", "Mårten Björkman" ]
Poster
Probabilistic Methods (eg, variational inference, causal inference, Gaussian processes)
Normalizing flows (NFs) have been shown to be advantageous in modeling complex distributions and improving sampling efficiency for unbiased sampling. In this work, we propose a new class of continuous NFs, ascent continuous normalizing flows (ACNFs), that makes a base distribution converge faster to a target distribut...
[ "normalizing flows", "gradient flows", "density estimation", "unbiased sampling", "variational inference" ]
null
6,593
null
null
[ 0.006682848557829857, -0.049066632986068726, 0.020379533991217613, 0.049232304096221924, 0.05169057846069336, 0.04426181688904762, 0.024093706160783768, -0.013782479800283909, -0.020243847742676735, -0.04866065829992294, 0.028291119262576103, -0.02602337673306465, -0.07262765616178513, 0.0...
Softened Symbol Grounding for Neuro-symbolic Systems
https://openreview.net/forum?id=HTJE5Krui0g
[ "Zenan Li", "Yuan Yao", "Taolue Chen", "Jingwei Xu", "Chun Cao", "Xiaoxing Ma", "Jian L\\\"{u}" ]
Poster
Deep Learning and representational learning
Neuro-symbolic learning generally consists of two separated worlds, i.e., neural network training and symbolic constraint solving, whose success hinges on symbol grounding, a fundamental problem in AI. This paper presents a novel, softened symbol grounding process, bridging the gap between the two worlds, and resultin...
[ "neuro-symbolic learning", "symbol grounding problem", "projection-based sampling" ]
null
6,551
2403.00323
title_snapshot
[ -0.040453046560287476, 0.0027556712739169598, -0.022306034341454506, 0.04009242728352547, 0.03374027833342552, 0.04572784900665283, 0.008981815539300442, -0.014966525137424469, -0.031321682035923004, -0.025916535407304764, -0.01638583280146122, 0.030924303457140923, -0.06257463246583939, -...
Encoding Recurrence into Transformers
https://openreview.net/forum?id=7YfHla7IxBJ
[ "Feiqing Huang", "Kexin Lu", "Yuxi CAI", "Zhen Qin", "Yanwen Fang", "Guangjian Tian", "Guodong Li" ]
Notable-top-5%
Deep Learning and representational learning
This paper novelly breaks down with ignorable loss an RNN layer into a sequence of simple RNNs, each of which can be further rewritten into a lightweight positional encoding matrix of a self-attention, named the Recurrence Encoding Matrix (REM). Thus, recurrent dynamics introduced by the RNN layer can be encapsulated i...
[ "Recurrent models", "Transformers", "sample efficiency", "gated mechanism" ]
null
6,550
null
null
[ 0.002796640619635582, -0.027998903766274452, -0.017681162804365158, 0.04008709266781807, 0.03488890081644058, 0.0560450442135334, 0.04002124443650246, 0.026721352711319923, -0.05671072006225586, -0.004284416325390339, -0.0008064987487159669, -0.0014477574732154608, -0.03444607928395271, -0...
Human-Guided Fair Classification for Natural Language Processing
https://openreview.net/forum?id=N_g8TT9Cy7f
[ "Florian E. Dorner", "Momchil Peychev", "Nikola Konstantinov", "Naman Goel", "Elliott Ash", "Martin Vechev" ]
Notable-top-25%
Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)
Text classifiers have promising applications in high-stake tasks such as resume screening and content moderation. These classifiers must be fair and avoid discriminatory decisions by being invariant to perturbations of sensitive attributes such as gender or ethnicity. However, there is a gap between human intuition abo...
[ "Individual Fairness", "Style Transfer", "NLP", "Crowdsourcing", "Human Evaluation" ]
null
6,549
2212.10154
title_snapshot
[ 0.0031269227620214224, -0.02989932894706726, -0.029904495924711227, 0.03794063255190849, 0.016725070774555206, 0.021264396607875824, 0.005280296318233013, 0.03606036305427551, -0.004573478829115629, -0.01110326498746872, -0.016632145270705223, 0.044106028974056244, -0.08227874338626862, -0...
Mini-batch $k$-means terminates within $O(d/\epsilon)$ iterations
https://openreview.net/forum?id=jREF4bkfi_S
[ "Gregory Schwartzman" ]
Poster
Theory (eg, control theory, learning theory, algorithmic game theory)
We answer the question: "Does \emph{local} progress (on batches) imply \emph{global} progress (on the entire dataset) for mini-batch $k$-means?". Specifically, we consider mini-batch $k$-means which terminates only when the improvement in the quality of the clustering on the sampled batch is below some threshold. Alth...
[]
null
6,542
2304.00419
title_snapshot
[ -0.014775908552110195, -0.03876620903611183, 0.012323545292019844, 0.012307762168347836, 0.024911969900131226, 0.05366174504160881, 0.046642135828733444, 0.014942655339837074, -0.020683225244283676, -0.021356334909796715, -0.011449920013546944, -0.04796510562300682, -0.04409830644726753, 0...
Learning Uncertainty for Unknown Domains with Zero-Target-Assumption
https://openreview.net/forum?id=pWVASryOyFw
[ "Yu Yu", "Hassan Sajjad", "Jia Xu" ]
Poster
Applications (eg, speech processing, computer vision, NLP)
We introduce our Maximum-Entropy Rewarded Reinforcement Learning (MERRL) framework that selects training data for more accurate Natural Language Processing (NLP). Because conventional data selection methods select training samples based on the test domain knowledge and not on real life data, they frequently fail in un...
[]
null
6,533
null
null
[ -0.01593487709760666, -0.01051847729831934, -0.0134300347417593, 0.05584249645471573, 0.024109605699777603, 0.028612004593014717, 0.024650797247886658, -0.013012240640819073, -0.006637361366301775, -0.02770581655204296, -0.06985323131084442, 0.05390804633498192, -0.06277979910373688, -0.01...
Transformer-based model for symbolic regression via joint supervised learning
https://openreview.net/forum?id=ULzyv9M1j5
[ "Wenqiang Li", "Weijun Li", "Linjun Sun", "Min Wu", "Lina Yu", "Jingyi Liu", "Yanjie Li", "Songsong Tian" ]
Poster
Applications (eg, speech processing, computer vision, NLP)
Symbolic regression (SR) is an important technique for discovering hidden mathematical expressions from observed data. Transformer-based approaches have been widely used for machine translation due to their high performance, and are recently highly expected to be used for SR. They input the data points, then output the...
[]
null
6,520
null
null
[ -0.027301602065563202, -0.041681595146656036, -0.029415979981422424, 0.009736493229866028, 0.04336460679769516, 0.06902904063463211, 0.03073422610759735, -0.019090812653303146, -0.016747374087572098, -0.015436983667314053, -0.021666334941983223, 0.04006672278046608, -0.059673987329006195, ...
QAID: Question Answering Inspired Few-shot Intent Detection
https://openreview.net/forum?id=gNI4_85Cyve
[ "Asaf Yehudai", "Matan Vetzler", "Yosi Mass", "Koren Lazar", "Doron Cohen", "Boaz Carmeli" ]
Poster
Deep Learning and representational learning
Intent detection with semantically similar fine-grained intents is a challenging task. To address it, we reformulate intent detection as a question-answering retrieval task by treating utterances and intent names as questions and answers. To that end, we utilize a question-answering retrieval architecture and adopt a t...
[ "Intent Detection", "Question Answering", "Contrastive Learning", "Passage Retrieval" ]
null
6,511
2303.01593
title_snapshot
[ 0.023309411481022835, -0.048694051802158356, -0.01393197849392891, 0.04317348822951317, 0.028255196288228035, 0.017960796132683754, 0.04122690483927727, -0.001974969869479537, 0.007879486307501793, 0.01361128594726324, -0.03836814686655998, 0.050433795899152756, -0.04523495212197304, -0.04...
Solving stochastic weak Minty variational inequalities without increasing batch size
https://openreview.net/forum?id=ejR4E1jaH9k
[ "Thomas Pethick", "Olivier Fercoq", "Puya Latafat", "Panagiotis Patrinos", "Volkan Cevher" ]
Poster
Optimization (eg, convex and non-convex optimization)
This paper introduces a family of stochastic extragradient-type algorithms for a class of nonconvex-nonconcave problems characterized by the weak Minty variational inequality (MVI). Unlike existing results on extragradient methods in the monotone setting, employing diminishing stepsizes is no longer possible in the wea...
[ "Variational inequalities", "stochastic first-order methods", "nonconvex-nonconcave", "minimax" ]
null
6,495
2302.09029
title_snapshot
[ -0.040038444101810455, 0.012334530241787434, -0.006639082450419664, 0.04505661502480507, 0.028215935453772545, 0.08128812909126282, 0.04029922932386398, -0.001263533835299313, -0.03838222473859787, -0.02587309293448925, -0.017259258776903152, -0.010330462828278542, -0.08224968612194061, -0...
Curriculum-based Co-design of Morphology and Control of Voxel-based Soft Robots
https://openreview.net/forum?id=r9fX833CsuN
[ "Yuxing Wang", "Shuang Wu", "Haobo Fu", "QIANG FU", "Tiantian Zhang", "Yongzhe Chang", "Xueqian Wang" ]
Poster
Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)
Co-design of morphology and control of a Voxel-based Soft Robot (VSR) is challenging due to the notorious bi-level optimization. In this paper, we present a Curriculum-based Co-design (CuCo) method for learning to design and control VSRs through an easy-to-difficult process. Specifically, we expand the design space fro...
[ "Artificial Life", "Brain-body Co-design", "Robotics", "Modular Soft Robots" ]
null
6,485
null
null
[ -0.01241062767803669, -0.020051492378115654, -0.011449952609837055, 0.04494588449597359, 0.03427281975746155, 0.023407142609357834, 0.0007990885060280561, -0.0033410941250622272, -0.03571771830320358, -0.03776827082037926, -0.023063937202095985, 0.011440354399383068, -0.050855379551649094, ...
WiNeRT: Towards Neural Ray Tracing for Wireless Channel Modelling and Differentiable Simulations
https://openreview.net/forum?id=tPKKXeW33YU
[ "Tribhuvanesh Orekondy", "Pratik Kumar", "Shreya Kadambi", "Hao Ye", "Joseph Soriaga", "Arash Behboodi" ]
Poster
Applications (eg, speech processing, computer vision, NLP)
In this paper, we work towards a neural surrogate to model wireless electro-magnetic propagation effects in indoor environments. Such neural surrogates provide a fast, differentiable, and continuous representation of the environment and enables end-to-end optimization for downstream tasks (e.g., network planning). Spec...
[ "neural rendering", "wireless", "ray tracing", "nerf" ]
null
6,480
null
null
[ -0.007716740481555462, -0.0038349346723407507, 0.014472121372818947, 0.009722174145281315, 0.05690338462591171, 0.01601015403866768, 0.016619782894849777, -0.00671561062335968, -0.01214691810309887, -0.08015388250350952, -0.0028984970413148403, 0.005616133101284504, -0.03435187414288521, -...
LS-IQ: Implicit Reward Regularization for Inverse Reinforcement Learning
https://openreview.net/forum?id=o3Q4m8jg4BR
[ "Firas Al-Hafez", "Davide Tateo", "Oleg Arenz", "Guoping Zhao", "Jan Peters" ]
Poster
Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)
Recent methods for imitation learning directly learn a $Q$-function using an implicit reward formulation rather than an explicit reward function. However, these methods generally require implicit reward regularization to improve stability and often mistreat absorbing states. Previous works show that a squared norm regu...
[ "Inverse Reinforcement Learning", "Imitation Learning", "Reward Regularization", "Deep Reinforcement Learning" ]
null
6,466
2303.00599
title_snapshot
[ -0.036808378994464874, -0.04107034206390381, -0.007307811174541712, 0.020488638430833817, 0.06288088858127594, 0.030701864510774612, 0.011733630672097206, 0.004436118062585592, -0.027789073064923286, -0.025949114933609962, -0.009805973619222641, 0.026797683909535408, -0.06771114468574524, ...
Humanly Certifying Superhuman Classifiers
https://openreview.net/forum?id=X5ZMzRYqUjB
[ "Qiongkai Xu", "Christian Walder", "Chenchen Xu" ]
Notable-top-25%
Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)
This paper addresses a key question in current machine learning research: if we believe that a model's predictions might be better than those given by human experts, how can we (humans) verify these beliefs? In some cases, this ``superhuman'' performance is readily demonstrated; for example by defeating top-tier human ...
[ "Evaluation theory", "Oracle accuracy", "Superhuman classifier" ]
null
6,465
2109.07867
title_snapshot
[ -0.03811696171760559, -0.007710753474384546, -0.029286446049809456, 0.03755056858062744, 0.04901386797428131, -0.00840599276125431, 0.023471005260944366, 0.011727078817784786, -0.016229689121246338, -0.0264222901314497, -0.039487965404987335, 0.020712461322546005, -0.06533388048410416, -0....
Share Your Representation Only: Guaranteed Improvement of the Privacy-Utility Tradeoff in Federated Learning
https://openreview.net/forum?id=oJpVVGXu9i
[ "Zebang Shen", "Jiayuan Ye", "Anmin Kang", "Hamed Hassani", "Reza Shokri" ]
Poster
Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)
Repeated parameter sharing in federated learning causes significant information leakage about private data, thus defeating its main purpose: data privacy. Mitigating the risk of this information leakage, using state of the art differentially private algorithms, also does not come for free. Randomized mechanisms can p...
[ "Differential Privacy", "Representation Learning", "Federated Learning" ]
null
6,462
2309.05505
title_snapshot
[ 0.005346141755580902, -0.025699432939291, -0.00025170494336634874, 0.08038785308599472, 0.04954526573419571, 0.02141929604113102, 0.02370399236679077, -0.039138905704021454, -0.01512485183775425, -0.04603532701730728, 0.00037166589754633605, -0.049882106482982635, -0.07199226319789886, 0.0...
EquiMod: An Equivariance Module to Improve Visual Instance Discrimination
https://openreview.net/forum?id=eDLwjKmtYFt
[ "Alexandre DEVILLERS", "Mathieu Lefort" ]
Poster
Deep Learning and representational learning
Recent self-supervised visual representation methods are closing the gap with supervised learning performance. Most of these successful methods rely on maximizing the similarity between embeddings of related synthetic inputs created through data augmentations. This can be seen as a task that encourages embeddings to le...
[ "Representation learning", "Self-supervised learning", "Contrastive learning", "Equivariance" ]
null
6,438
2211.01244
title_judge
[ 0.05885966494679451, -0.016672268509864807, -0.007716714404523373, 0.031269535422325134, 0.03232405334711075, 0.032966192811727524, 0.017029860988259315, 0.007903438061475754, -0.05936861410737038, -0.03811027109622955, -0.0352323018014431, -0.00961090624332428, -0.09682392328977585, -0.00...
Task-Aware Information Routing from Common Representation Space in Lifelong Learning
https://openreview.net/forum?id=-M0TNnyWFT5
[ "Prashant Shivaram Bhat", "Bahram Zonooz", "Elahe Arani" ]
Poster
Deep Learning and representational learning
Intelligent systems deployed in the real world suffer from catastrophic forgetting when exposed to a sequence of tasks. Humans, on the other hand, acquire, consolidate, and transfer knowledge between tasks that rarely interfere with the consolidated knowledge. Accompanied by self-regulated neurogenesis, continual lear...
[ "Continual learning", "Lifelong learning", "Representation learning", "Global workspace theory", "Task-specific attention" ]
null
6,431
2302.11346
title_snapshot
[ 0.0031048220116645098, 0.0008109945920296013, -0.032306864857673645, 0.01557182241231203, 0.04689861834049225, 0.01849260926246643, 0.03911057487130165, 0.029457466676831245, -0.0484020821750164, -0.023411797359585762, -0.020310737192630768, 0.0010421497281640768, -0.054356273263692856, -0...
CodeBPE: Investigating Subtokenization Options for Large Language Model Pretraining on Source Code
https://openreview.net/forum?id=htL4UZ344nF
[ "Nadezhda Chirkova", "Sergey Troshin" ]
Poster
Applications (eg, speech processing, computer vision, NLP)
Recent works have widely adopted large language model pretraining for source code, suggested source code-specific pretraining objectives and investigated the applicability of various Transformer-based language model architectures for source code. This work investigates another important aspect of such models, the effec...
[ "source code processing", "tokenization", "byte-pair encoding" ]
null
6,429
2308.00683
title_snapshot
[ -0.005233630537986755, -0.06644809246063232, -0.02728739194571972, 0.003997389692813158, 0.032490868121385574, 0.0498298779129982, 0.02451850287616253, -0.0067131672985851765, -0.008727703243494034, -0.028078550472855568, -0.03152437508106232, 0.02026163600385189, -0.06197807192802429, -0....
Few-Shot Domain Adaptation For End-to-End Communication
https://openreview.net/forum?id=4F1gvduDeL
[ "Jayaram Raghuram", "Yijing Zeng", "Dolores Garcia", "Rafael Ruiz", "Somesh Jha", "Joerg Widmer", "Suman Banerjee" ]
Notable-top-25%
Applications (eg, speech processing, computer vision, NLP)
The problem of end-to-end learning of a communication system using an autoencoder -- consisting of an encoder, channel, and decoder modeled using neural networks -- has recently been shown to be an effective approach. A challenge faced in the practical adoption of this learning approach is that under changing channel c...
[ "domain adaptation", "end-to-end communication", "autoencoders", "Gaussian mixtures", "mixture density networks", "few-shot", "wireless channel" ]
null
6,412
2108.00874
title_snapshot
[ 0.0014920660760253668, -0.028604209423065186, -0.017476631328463554, 0.050644468516111374, 0.06909225136041641, 0.02486036904156208, 0.052081845700740814, -0.0191132053732872, 0.003872646950185299, -0.05127960070967674, -0.022687844932079315, 0.0410885214805603, -0.06013898551464081, 0.004...
FairGBM: Gradient Boosting with Fairness Constraints
https://openreview.net/forum?id=x-mXzBgCX3a
[ "André Cruz", "Catarina G Belém", "João Bravo", "Pedro Saleiro", "Pedro Bizarro" ]
Poster
Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)
Tabular data is prevalent in many high-stakes domains, such as financial services or public policy. Gradient Boosted Decision Trees (GBDT) are popular in these settings due to their scalability, performance, and low training cost. While fairness in these domains is a foremost concern, existing in-processing Fair ML met...
[ "fairness", "gradient boosting", "constrained optimization", "tabular data" ]
null
6,390
2209.07850
title_snapshot
[ -0.023425044491887093, -0.05055586248636246, -0.014089255593717098, 0.04313512146472931, -0.0035110379103571177, 0.022041022777557373, 0.015509209595620632, 0.012330067344009876, -0.042629167437553406, -0.011700290255248547, 0.006213263608515263, 0.013146976940333843, -0.09362295269966125, ...
Online Bias Correction for Task-Free Continual Learning
https://openreview.net/forum?id=18XzeuYZh_
[ "Aristotelis Chrysakis", "Marie-Francine Moens" ]
Poster
General Machine Learning (ie none of the above)
Task-free continual learning is the machine-learning setting where a model is trained online with data generated by a nonstationary stream. Conventional wisdom suggests that, in this setting, models are trained using an approach called experience replay, where the risk is computed both with respect to current stream ob...
[ "Task-Free Continual Learning" ]
null
6,379
null
null
[ 0.0035302455071359873, -0.018488675355911255, -0.006173381581902504, 0.016383353620767593, 0.04900787025690079, 0.022322827950119972, 0.023654231801629066, 0.03823074698448181, -0.04072044417262077, -0.03907942771911621, -0.018409550189971924, 0.029302485287189484, -0.07539549469947815, -0...
Don’t fear the unlabelled: safe semi-supervised learning via debiasing
https://openreview.net/forum?id=TN9gQ4x0Ep3
[ "Hugo Schmutz", "Olivier HUMBERT", "Pierre-Alexandre Mattei" ]
Poster
Unsupervised and Self-supervised learning
Semi-supervised learning (SSL) provides an effective means of leveraging unlabelled data to improve a model’s performance. Even though the domain has received a considerable amount of attention in the past years, most methods present the common drawback of lacking theoretical guarantees. Our starting point is to notice...
[ "Semi-supervised learning", "deep learning", "empirical risk minimisation", "control variate", "variance reduction", "asymptotic statistics" ]
null
6,362
2203.07512
title_judge
[ 0.013973946683108807, -0.05876178294420242, -0.02776288613677025, 0.05137612298130989, 0.028895577415823936, -0.01399050373584032, 0.037616293877363205, -0.01130647398531437, -0.010737470351159573, -0.04331902787089348, -0.02386072836816311, 0.023388702422380447, -0.089263416826725, 0.0054...
Learning a Data-Driven Policy Network for Pre-Training Automated Feature Engineering
https://openreview.net/forum?id=688hNNMigVX
[ "Liyao Li", "Haobo Wang", "Liangyu Zha", "Qingyi Huang", "Sai Wu", "Gang Chen", "Junbo Zhao" ]
Notable-top-25%
General Machine Learning (ie none of the above)
Feature engineering is widely acknowledged to be pivotal in tabular data analysis and prediction. Automated feature engineering (AutoFE) emerged to automate this process managed by experienced data scientists and engineers conventionally. In this area, most — if not all — prior work adopted an identical framework from ...
[ "Automated Feature Engineering", "Reinforcement Learning", "Tabular Data", "Data-Driven", "Pre-Training" ]
null
6,357
null
null
[ -0.03533236309885979, -0.039041854441165924, -0.019265078008174896, 0.04053781181573868, 0.07807990908622742, 0.036842893809080124, 0.013903387822210789, -0.051462315022945404, -0.005512757692486048, -0.04181912913918495, -0.021155662834644318, 0.0009562769555486739, -0.06616996973752975, ...
Making Substitute Models More Bayesian Can Enhance Transferability of Adversarial Examples
https://openreview.net/forum?id=bjPPypbLre
[ "Qizhang Li", "Yiwen Guo", "Wangmeng Zuo", "Hao Chen" ]
Poster
Deep Learning and representational learning
The transferability of adversarial examples across deep neural networks (DNNs) is the crux of many black-box attacks. Many prior efforts have been devoted to improving the transferability via increasing the diversity in inputs of some substitute models. In this paper, by contrast, we opt for the diversity in substitute...
[ "Adversarial Examples", "Black-box Attacks", "Adversarial Transferability" ]
null
6,351
2302.05086
title_snapshot
[ 0.0034802004229277372, -0.022453822195529938, -0.022334489971399307, 0.05759696289896965, 0.037797700613737106, 0.01652509719133377, 0.022940102964639664, -0.02490444853901863, -0.004855051636695862, -0.05358421057462692, -0.0014467270812019706, 0.012747221626341343, -0.05307307094335556, ...
Learning Group Importance using the Differentiable Hypergeometric Distribution
https://openreview.net/forum?id=75O7S_L4oY
[ "Thomas M. Sutter", "Laura Manduchi", "Alain Ryser", "Julia E Vogt" ]
Notable-top-25%
Deep Learning and representational learning
Partitioning a set of elements into subsets of a priori unknown sizes is essential in many applications. These subset sizes are rarely explicitly learned - be it the cluster sizes in clustering applications or the number of shared versus independent generative latent factors in weakly-supervised learning. Probability d...
[ "hypergeometric distribution", "weakly-supervised learning", "reparameterization trick", "group importance", "variational clustering", "gumbel softmax" ]
null
6,350
2203.01629
title_snapshot
[ -0.008653838187456131, -0.007285722065716982, 0.02696160040795803, 0.033841464668512344, 0.03437541052699089, 0.051281217485666275, 0.01530587486922741, -0.033702731132507324, -0.024262426421046257, -0.032788943499326706, -0.00473402626812458, -0.01023799180984497, -0.07380038499832153, 0....
Cross-Layer Retrospective Retrieving via Layer Attention
https://openreview.net/forum?id=pvgEL1yS3Ql
[ "Yanwen Fang", "Yuxi CAI", "Jintai Chen", "Jingyu Zhao", "Guangjian Tian", "Guodong Li" ]
Poster
Deep Learning and representational learning
More and more evidence has shown that strengthening layer interactions can enhance the representation power of a deep neural network, while self-attention excels at learning interdependencies by retrieving query-activated information. Motivated by this, we devise a cross-layer attention mechanism, called multi-head rec...
[ "Layer Attention", "Recurrent Layer Attention", "Layer Interaction", "CNNs", "Vision Transformers", "Vision Networks" ]
null
6,346
2302.03985
title_snapshot
[ 0.002757095033302903, -0.010206181555986404, -0.000820724933873862, 0.02046215534210205, 0.015163645148277283, 0.01901453174650669, 0.005619359202682972, 0.007279491517692804, -0.040106456726789474, -0.016924740746617317, -0.016317540779709816, -0.0018899162532761693, -0.05693921074271202, ...
Decision S4: Efficient Sequence-Based RL via State Spaces Layers
https://openreview.net/forum?id=kqHkCVS7wbj
[ "Shmuel Bar David", "Itamar Zimerman", "Eliya Nachmani", "Lior Wolf" ]
Poster
Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)
Recently, sequence learning methods have been applied to the problem of off-policy Reinforcement Learning, including the seminal work on Decision Transformers, which employs transformers for this task. Since transformers are parameter-heavy, cannot benefit from history longer than a fixed window size, and are not compu...
[ "Sequential RL", "S4", "Decision transformers" ]
null
6,339
2306.05167
title_snapshot
[ -0.0351293571293354, -0.045080918818712234, -0.002143049845471978, 0.03217891976237297, 0.05813050642609596, 0.023571962490677834, -0.0015295022167265415, 0.008520773611962795, -0.02975151129066944, -0.013541791588068008, -0.013087506406009197, 0.04239385575056076, -0.07912030816078186, -0...
Unveiling the sampling density in non-uniform geometric graphs
https://openreview.net/forum?id=mnVf1W6ipGm
[ "Raffaele Paolino", "Aleksandar Bojchevski", "Stephan Günnemann", "Gitta Kutyniok", "Ron Levie" ]
Poster
Deep Learning and representational learning
A powerful framework for studying graphs is to consider them as geometric graphs: nodes are randomly sampled from an underlying metric space, and any pair of nodes is connected if their distance is less than a specified neighborhood radius. Currently, the literature mostly focuses on uniform sampling and constant neigh...
[ "graph neural network", "graph representation learning", "spectral method", "non-uniform sampling", "geometric graph", "graphon" ]
null
6,335
2210.08219
title_snapshot
[ 0.014867688529193401, -0.030834443867206573, -0.00957784429192543, 0.0621785968542099, 0.037471622228622437, 0.021031947806477547, 0.015407891944050789, 0.01938064955174923, 0.0030375465285032988, -0.09034450352191925, 0.0010728581110015512, -0.04849771782755852, -0.07465696334838867, 0.00...
Boosting Causal Discovery via Adaptive Sample Reweighting
https://openreview.net/forum?id=LNpMtk15AS4
[ "An Zhang", "Fangfu Liu", "Wenchang Ma", "Zhibo Cai", "Xiang Wang", "Tat-Seng Chua" ]
Poster
General Machine Learning (ie none of the above)
Under stringent model type and variable distribution assumptions, score-based causal discovery methods learn the directed acyclic graph (DAG) from observational data by evaluating candidate graphs over an averaged score function. Despite the great success in low-dimensional linear systems, it has been observed that the...
[ "Causal Structure Learning", "Score-based Causal Discovery", "Adaptive Sample Reweighting" ]
null
6,333
2303.03187
title_judge
[ -0.005168410018086433, -0.0347769558429718, 0.0006656326586380601, 0.04863692820072174, 0.031832773238420486, 0.022420866414904594, 0.03632648289203644, -0.016920028254389763, 0.004138884600251913, -0.050452351570129395, 0.021830683574080467, 0.00683671934530139, -0.07002969086170197, 0.00...
Iterative Circuit Repair Against Formal Specifications
https://openreview.net/forum?id=SEcSahl0Ql
[ "Matthias Cosler", "Frederik Schmitt", "Christopher Hahn", "Bernd Finkbeiner" ]
Poster
Applications (eg, speech processing, computer vision, NLP)
We present a deep learning approach for repairing sequential circuits against formal specifications given in linear-time temporal logic (LTL). Given a defective circuit and its formal specification, we train Transformer models to output circuits that satisfy the corresponding specification. We propose a separated hiera...
[ "sequential circuits", "repair", "synthesis", "transformer" ]
null
6,328
2303.01158
title_snapshot
[ 0.00386873004026711, -0.015952011570334435, -0.02166232466697693, 0.0812106728553772, 0.04372907802462578, 0.035639505833387375, 0.004492415580898523, 0.01179504208266735, 0.006416006479412317, 0.01587298884987831, 0.03899088501930237, 0.026359660550951958, -0.04489436373114586, 0.01065138...
Can BERT Refrain from Forgetting on Sequential Tasks? A Probing Study
https://openreview.net/forum?id=UazgYBMS9-W
[ "Mingxu Tao", "Yansong Feng", "Dongyan Zhao" ]
Poster
Applications (eg, speech processing, computer vision, NLP)
Large pre-trained language models have helped to achieve state of the art on a variety of NLP tasks, nevertheless, they still suffer from forgetting when incrementally learning a series of sequential tasks. To alleviate this problem, recent works propose several models enhanced by sparse experience replay and local ada...
[ "Natural Language Processing", "Probing Study" ]
null
6,325
2303.01081
title_snapshot
[ -0.029040949419140816, -0.01464133895933628, -0.0009298911318182945, 0.04094014689326286, 0.0319133959710598, -0.0020409126300364733, 0.03515353053808212, 0.03274824097752571, -0.03759577497839928, -0.009116764180362225, -0.008306322619318962, 0.03871756047010422, -0.0350772924721241, 0.00...
Behavior Proximal Policy Optimization
https://openreview.net/forum?id=3c13LptpIph
[ "Zifeng Zhuang", "Kun LEI", "Jinxin Liu", "Donglin Wang", "Yilang Guo" ]
Poster
Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)
Offline reinforcement learning (RL) is a challenging setting where existing off-policy actor-critic methods perform poorly due to overestimating of out-of-distribution state-action pairs. Thus, various additional augmentations are proposed to keep the learned policy close to the offline dataset (or the behavior policy)...
[ "Offline Reinforcement Learning", "Monotonic Policy Improvement" ]
null
6,324
2302.11312
title_snapshot
[ -0.03520854562520981, -0.03978604078292847, 0.013066786341369152, 0.04588599503040314, 0.04829363897442818, 0.018984096124768257, 0.023897305130958557, -0.0008266135700978339, -0.045415088534355164, -0.015610422939062119, -0.033063989132642746, 0.008640510030090809, -0.0866025760769844, -0...
Actionable Neural Representations: Grid Cells from Minimal Constraints
https://openreview.net/forum?id=xfqDe72zh41
[ "Will Dorrell", "Peter E. Latham", "Timothy E. J. Behrens", "James C. R. Whittington" ]
Poster
Neuroscience and Cognitive Science (e.g., neural coding, brain-computer interfaces)
To afford flexible behaviour, the brain must build internal representations that mirror the structure of variables in the external world. For example, 2D space obeys rules: the same set of actions combine in the same way everywhere (step north, then south, and you won't have moved, wherever you start). We suggest the b...
[ "Grid Cells", "Representation Theory", "Theoretical Neuroscience", "Normative Models" ]
null
6,321
2209.15563
title_snapshot
[ -0.032133884727954865, -0.0026306351646780968, -0.013771998696029186, 0.014828024432063103, 0.01716715469956398, 0.010569709353148937, 0.013609716668725014, 0.018730876967310905, -0.04618360847234726, -0.045851752161979675, -0.02074115350842476, -0.041395656764507294, -0.06307073682546616, ...
Modeling content creator incentives on algorithm-curated platforms
https://openreview.net/forum?id=l6CpxixmUg
[ "Jiri Hron", "Karl Krauth", "Michael Jordan", "Niki Kilbertus", "Sarah Dean" ]
Notable-top-5%
Theory (eg, control theory, learning theory, algorithmic game theory)
Content creators compete for user attention. Their reach crucially depends on algorithmic choices made by developers on online platforms. To maximize exposure, many creators adapt strategically, as evidenced by examples like the sprawling search engine optimization industry. This begets competition for the finite user ...
[ "Nash equilibria", "producer incentives", "attention monetizing platforms", "recommenders", "differentiable games", "exposure game" ]
null
6,317
2206.13102
title_snapshot
[ 0.005692870356142521, -0.024722304195165634, 0.005638757720589638, 0.03271624073386192, 0.029080277308821678, 0.01928587444126606, 0.015611954964697361, 0.025450648739933968, -0.033204395323991776, 0.008149084635078907, -0.021817008033394814, 0.007305149920284748, -0.03874586895108223, -0....
Mole-BERT: Rethinking Pre-training Graph Neural Networks for Molecules
https://openreview.net/forum?id=jevY-DtiZTR
[ "Jun Xia", "Chengshuai Zhao", "Bozhen Hu", "Zhangyang Gao", "Cheng Tan", "Yue Liu", "Siyuan Li", "Stan Z. Li" ]
Poster
Deep Learning and representational learning
Recent years have witnessed the prosperity of pre-training graph neural networks (GNNs) for molecules. Typically, atom types as node attributes are randomly masked, and GNNs are then trained to predict masked types as in AttrMask \citep{hu2020strategies}, following the Masked Language Modeling (MLM) task of BERT~\citep...
[ "graph neural networks" ]
null
6,308
null
null
[ -0.020775167271494865, 0.03350331261754036, 0.006499054841697216, 0.06322138011455536, 0.03251294791698456, 0.009114491753280163, 0.003950390499085188, 0.000269530137302354, 0.0101914182305336, -0.02885645255446434, 0.006710100453346968, 0.030606862157583237, -0.06473930925130844, 0.010047...
Concept-level Debugging of Part-Prototype Networks
https://openreview.net/forum?id=oiwXWPDTyNk
[ "Andrea Bontempelli", "Stefano Teso", "Katya Tentori", "Fausto Giunchiglia", "Andrea Passerini" ]
Notable-top-25%
Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)
Part-prototype Networks (ProtoPNets) are concept-based classifiers designed to achieve the same performance as black-box models without compromising transparency. ProtoPNets compute predictions based on similarity to class-specific part-prototypes learned to recognize parts of training examples, making it easy to faith...
[ "explainability", "debugging", "self-explainable networks", "part-prototype networks", "concept-based models" ]
null
6,296
2205.15769
title_snapshot
[ -0.03230720013380051, -0.027533629909157753, -0.03906622901558876, 0.04539753496646881, 0.041307948529720306, 0.006408018991351128, 0.016964931041002274, -0.013021496124565601, 0.001277124392800033, -0.04612838104367256, -0.012508036568760872, 0.000725378340575844, -0.05108647793531418, 0....
Geometrically regularized autoencoders for non-Euclidean data
https://openreview.net/forum?id=_q7A0m3vXH0
[ "Cheongjae Jang", "Yonghyeon Lee", "Yung-Kyun Noh", "Frank C. Park" ]
Poster
Deep Learning and representational learning
Regularization is almost {\it de rigueur} when designing autoencoders that are sparse and robust to noise. Given the recent surge of interest in machine learning problems involving non-Euclidean data, in this paper we address the regularization of autoencoders on curved spaces. We show that by ignoring the underlying g...
[ "autoencoders", "Riemannian geometry", "non-Euclidean data", "regularization", "score estimation" ]
null
6,294
null
null
[ 0.016202135011553764, -0.04080559313297272, 0.0024616383016109467, 0.05151432380080223, 0.04725690558552742, 0.06655827909708023, 0.024740025401115417, 0.008626556023955345, -0.02339976653456688, -0.08000613003969193, -0.022459836676716805, -0.004113053437322378, -0.06283678114414215, 0.01...
A Message Passing Perspective on Learning Dynamics of Contrastive Learning
https://openreview.net/forum?id=VBTJqqWjxMv
[ "Yifei Wang", "Qi Zhang", "Tianqi Du", "Jiansheng Yang", "Zhouchen Lin", "Yisen Wang" ]
Poster
Unsupervised and Self-supervised learning
In recent years, contrastive learning achieves impressive results on self-supervised visual representation learning, but there still lacks a rigorous understanding of its learning dynamics. In this paper, we show that if we cast a contrastive objective equivalently into the feature space, then its learning dynamics adm...
[]
null
6,281
2303.04435
title_snapshot
[ -0.007773103658109903, -0.021533554419875145, -0.015634849667549133, 0.05138498544692993, 0.01929754950106144, 0.03349461406469345, 0.040460433810949326, 0.01671779900789261, -0.04780399799346924, -0.011832751333713531, 0.004028022289276123, -0.010628190822899342, -0.07938406616449356, 0.0...
Zeroth-Order Optimization with Trajectory-Informed Derivative Estimation
https://openreview.net/forum?id=n1bLgxHW6jW
[ "Yao Shu", "Zhongxiang Dai", "Weicong Sng", "Arun Verma", "Patrick Jaillet", "Bryan Kian Hsiang Low" ]
Poster
Optimization (eg, convex and non-convex optimization)
Zeroth-order (ZO) optimization, in which the derivative is unavailable, has recently succeeded in many important machine learning applications. Existing algorithms rely on finite difference (FD) methods for derivative estimation and gradient descent (GD)-based approaches for optimization. However, these algorithms suff...
[ "zeroth-order optimization", "derivative estimation", "finite difference" ]
null
6,273
null
null
[ -0.026984628289937973, -0.007332842331379652, 0.012728058733046055, 0.052861202508211136, 0.029709484428167343, 0.04855642095208168, 0.03713032230734825, 0.021489804610610008, -0.009786814451217651, -0.01307023037225008, 0.002059189835563302, 0.024396056309342384, -0.04669995233416557, 0.0...
Neuroevolution is a Competitive Alternative to Reinforcement Learning for Skill Discovery
https://openreview.net/forum?id=6BHlZgyPOZY
[ "Felix Chalumeau", "Raphael Boige", "Bryan Lim", "Valentin Macé", "Maxime Allard", "Arthur Flajolet", "Antoine Cully", "Thomas PIERROT" ]
Notable-top-25%
Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)
Deep Reinforcement Learning (RL) has emerged as a powerful paradigm for training neural policies to solve complex control tasks. However, these policies tend to be overfit to the exact specifications of the task and environment they were trained on, and thus do not perform well when conditions deviate slightly or when ...
[]
null
6,262
2210.03516
title_snapshot
[ -0.04466382414102554, -0.04386014863848686, -0.030227139592170715, 0.03303230553865433, 0.057142335921525955, 0.03977656736969948, 0.029080986976623535, -0.02504531480371952, -0.03485359251499176, -0.04597512260079384, -0.0030534989200532436, 0.021996282041072845, -0.05928884446620941, -0....
Uniform-in-time propagation of chaos for the mean-field gradient Langevin dynamics
https://openreview.net/forum?id=_JScUk9TBUn
[ "Taiji Suzuki", "Atsushi Nitanda", "Denny Wu" ]
Poster
Theory (eg, control theory, learning theory, algorithmic game theory)
The mean-field Langevin dynamics is characterized by a stochastic differential equation that arises from (noisy) gradient descent on an infinite-width two-layer neural network, which can be viewed as an interacting particle system. In this work, we establish a quantitative weak propagation of chaos result for the syste...
[ "Neural network optimization", "mean-field regime", "interacting particle system", "propagation of chaos" ]
null
6,257
null
null
[ -0.026836007833480835, -0.0017250124365091324, 0.0030896991956979036, 0.01806262880563736, 0.05189787968993187, 0.023771114647388458, 0.027402037754654884, 0.003973004408180714, -0.032055728137493134, -0.05024208873510361, 0.025360187515616417, -0.0009857014520093799, -0.0563124343752861, ...
Asynchronous Distributed Bilevel Optimization
https://openreview.net/forum?id=_i0-12XqVJZ
[ "Yang Jiao", "Kai Yang", "Tiancheng Wu", "Dongjin Song", "Chengtao Jian" ]
Poster
Optimization (eg, convex and non-convex optimization)
Bilevel optimization plays an essential role in many machine learning tasks, ranging from hyperparameter optimization to meta-learning. Existing studies on bilevel optimization, however, focus on either centralized or synchronous distributed setting. The centralized bilevel optimization approaches require collecting ma...
[]
null
6,251
2212.10048
title_snapshot
[ -0.044771257787942886, -0.02764941193163395, 0.020848138257861137, 0.043560691177845, 0.011172231286764145, 0.023443523794412613, 0.01619854010641575, -0.015354319475591183, -0.030871741473674774, -0.05987773835659027, 0.013033001683652401, -0.02600155957043171, -0.0599459633231163, -0.004...
Confidence-Based Feature Imputation for Graphs with Partially Known Features
https://openreview.net/forum?id=YPKBIILy-Kt
[ "Daeho Um", "Jiwoong Park", "Seulki Park", "Jin young Choi" ]
Poster
Deep Learning and representational learning
This paper investigates a missing feature imputation problem for graph learning tasks. Several methods have previously addressed learning tasks on graphs with missing features. However, in cases of high rates of missing features, they were unable to avoid significant performance degradation. To overcome this limitation...
[ "Graph neural networks", "Graphs", "Missing features" ]
null
6,247
2305.16618
title_snapshot
[ -0.01737675629556179, -0.043428823351860046, -0.010772837325930595, 0.08615418523550034, 0.061522334814071655, 0.038688719272613525, 0.05875604227185249, 0.004881615750491619, 0.018573885783553123, -0.06333953142166138, -0.00663834810256958, -0.007254228461533785, -0.06923288851976395, 0.0...
LiftedCL: Lifting Contrastive Learning for Human-Centric Perception
https://openreview.net/forum?id=WHlt5tLz12T
[ "Ziwei Chen", "Qiang Li", "Xiaofeng Wang", "Wankou Yang" ]
Poster
Unsupervised and Self-supervised learning
Human-centric perception targets for understanding human body pose, shape and segmentation. Pre-training the model on large-scale datasets and fine-tuning it on specific tasks has become a well-established paradigm in human-centric perception. Recently, self-supervised learning methods have re-investigated contrastive ...
[ "contrastive learning", "human-centric perception" ]
null
6,237
null
null
[ 0.00835469365119934, -0.01169097051024437, -0.009042453020811081, 0.011057027615606785, 0.028968244791030884, 0.0025361166335642338, 0.017825720831751823, 0.03283616900444031, -0.05072498321533203, -0.021908868104219437, -0.03950183838605881, -0.009387240745127201, -0.07769263535737991, -0...
Individual Privacy Accounting with Gaussian Differential Privacy
https://openreview.net/forum?id=JmC_Tld3v-f
[ "Antti Koskela", "Marlon Tobaben", "Antti Honkela" ]
Poster
Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)
Individual privacy accounting enables bounding differential privacy (DP) loss individually for each participant involved in the analysis. This can be informative as often the individual privacy losses are considerably smaller than those indicated by the DP bounds that are based on considering worst-case bounds at each ...
[ "differential privacy", "gaussian differential privacy", "fully adaptive compositions", "privacy accounting", "individual privacy loss" ]
null
6,236
2209.15596
title_snapshot
[ 0.0014984585577622056, 0.03351423516869545, 0.006027251482009888, 0.045341577380895615, 0.049103155732154846, 0.02845103293657303, 0.03595122694969177, -0.028707878664135933, -0.016240453347563744, -0.031066065654158592, 0.004180707037448883, -0.01235579140484333, -0.03980157896876335, -0....
Evolving Populations of Diverse RL Agents with MAP-Elites
https://openreview.net/forum?id=CBfYffLqWqb
[ "Thomas PIERROT", "Arthur Flajolet" ]
Poster
Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)
Quality Diversity (QD) has emerged as a powerful alternative optimization paradigm that aims at generating large and diverse collections of solutions, notably with its flagship algorithm MAP-ELITES (ME) which evolves solutions through mutations and crossovers. While very effective for some unstructured problems, early ...
[]
null
6,235
2303.12803
title_snapshot
[ -0.04073905199766159, -0.037370651960372925, 0.002942413790151477, 0.04691378027200699, 0.04077412188053131, 0.042049769312143326, 0.02770034596323967, -0.010855999775230885, -0.04730624333024025, -0.05064203217625618, -0.023911647498607635, 0.013168944977223873, -0.06770859658718109, -0.0...
Implicit Bias in Leaky ReLU Networks Trained on High-Dimensional Data
https://openreview.net/forum?id=JpbLyEI5EwW
[ "Spencer Frei", "Gal Vardi", "Peter Bartlett", "Nathan Srebro", "Wei Hu" ]
Notable-top-25%
Deep Learning and representational learning
The implicit biases of gradient-based optimization algorithms are conjectured to be a major factor in the success of modern deep learning. In this work, we investigate the implicit bias of gradient flow and gradient descent in two-layer fully-connected neural networks with leaky ReLU activations when the training data...
[ "implicit bias", "gradient descent", "gradient flow", "neural networks" ]
null
6,233
2210.07082
title_snapshot
[ -0.03276585415005684, 0.0017246071947738528, -0.00015119252202566713, 0.036919523030519485, 0.01024720910936594, 0.025036940351128578, 0.026010936126112938, 0.0028537041507661343, -0.0035112695768475533, -0.022948767989873886, -0.002529953373596072, -0.012020685710012913, -0.0667320489883422...
Gray-Box Gaussian Processes for Automated Reinforcement Learning
https://openreview.net/forum?id=rmoMvptXK7M
[ "Gresa Shala", "André Biedenkapp", "Frank Hutter", "Josif Grabocka" ]
Poster
Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)
Despite having achieved spectacular milestones in an array of important real-world applications, most Reinforcement Learning (RL) methods are very brittle concerning their hyperparameters. Notwithstanding the crucial importance of setting the hyperparameters in training state-of-the-art agents, the task of hyperparamet...
[]
null
6,226
null
null
[ -0.01855044811964035, 0.0035277153365314007, -0.0038444336969405413, 0.04959084466099739, 0.015509854070842266, 0.05391181632876396, 0.001144822919741273, -0.005943290423601866, -0.01695922203361988, -0.03329094871878624, -0.018457535654306412, 0.027889149263501167, -0.07015880942344666, -...
Protein Sequence and Structure Co-Design with Equivariant Translation
https://openreview.net/forum?id=pRCMXcfdihq
[ "Chence Shi", "Chuanrui Wang", "Jiarui Lu", "Bozitao Zhong", "Jian Tang" ]
Poster
Machine Learning for Sciences (eg biology, physics, health sciences, social sciences, climate/sustainability )
Proteins are macromolecules that perform essential functions in all living organisms. Designing novel proteins with specific structures and desired functions has been a long-standing challenge in the field of bioengineering. Existing approaches generate both protein sequence and structure using either autoregressive mo...
[ "protein design", "sequence structure co-design", "equivariant translation", "geometric deep learning" ]
null
6,221
2210.08761
title_snapshot
[ 0.018235914409160614, 0.011917642317712307, -0.014160108752548695, 0.032432932406663895, 0.030441254377365112, 0.015801655128598213, 0.009917641058564186, 0.00836990587413311, 0.012288959696888924, -0.03807167708873749, 0.028901146724820137, -0.02463223785161972, -0.04489994794130325, 0.01...
Learning in temporally structured environments
https://openreview.net/forum?id=z0_V5O9cmNw
[ "Matt Jones", "Tyler R. Scott", "Mengye Ren", "Gamaleldin Fathy Elsayed", "Katherine Hermann", "David Mayo", "Michael Curtis Mozer" ]
Poster
Theory (eg, control theory, learning theory, algorithmic game theory)
Natural environments have temporal structure at multiple timescales. This property is reflected in biological learning and memory but typically not in machine learning systems. We advance a multiscale learning method in which each weight in a neural network is decomposed as a sum of subweights with different learning a...
[ "1/f noise", "Kalman filter", "neural network", "learning theory", "optimizers" ]
null
6,215
null
null
[ -0.008602445013821125, 0.005036582238972187, 0.015688853338360786, 0.0422762855887413, 0.04425092041492462, 0.032841864973306656, 0.02287084236741066, 0.00857166014611721, -0.018513107672333717, -0.0327785424888134, 0.02849254570901394, 0.010036860592663288, -0.056949932128190994, 0.000295...
RandProx: Primal-Dual Optimization Algorithms with Randomized Proximal Updates
https://openreview.net/forum?id=cB4N3G5udUS
[ "Laurent Condat", "Peter Richtárik" ]
Poster
Optimization (eg, convex and non-convex optimization)
Proximal splitting algorithms are well suited to solving large-scale nonsmooth optimization problems, in particular those arising in machine learning. We propose a new primal–dual algorithm, in which the dual update is randomized; equivalently, the proximity operator of one of the function in the problem is replaced by...
[ "optimization", "randomized algorithm", "stochastic algorithm", "proximal splitting", "proximity operator" ]
null
6,212
2207.12891
title_snapshot
[ -0.01925421878695488, -0.01013867650181055, -0.0041534085758030415, 0.054204996675252914, 0.039836302399635315, 0.052821896970272064, 0.002325284993276, -0.003964823670685291, -0.031667426228523254, -0.06028240919113159, -0.019321639090776443, -0.003713698126375675, -0.05353270098567009, -...
Preserving Pre-trained Features Helps Calibrate Fine-tuned Language Models
https://openreview.net/forum?id=NI7StoWHJPT
[ "Guande He", "Jianfei Chen", "Jun Zhu" ]
Poster
Deep Learning and representational learning
Large pre-trained language models (PLMs) have demonstrated strong performance on natural language understanding (NLU) tasks through fine-tuning. However, fine-tuned models still suffer from overconfident predictions, especially in out-of-domain settings. In this paper, we tackle the problem of calibrating fine-tuned la...
[]
null
6,210
2305.19249
title_snapshot
[ -0.028786852955818176, -0.017685621976852417, 0.0052865780889987946, 0.030829042196273804, 0.0561511367559433, 0.04551215097308159, 0.04007809981703758, 0.007717822212725878, -0.034306321293115616, -0.010355236008763313, 0.0017459650989621878, 0.03083939664065838, -0.03508523851633072, 0.0...
Guarded Policy Optimization with Imperfect Online Demonstrations
https://openreview.net/forum?id=O5rKg7IRQIO
[ "Zhenghai Xue", "Zhenghao Peng", "Quanyi Li", "Zhihan Liu", "Bolei Zhou" ]
Notable-top-25%
Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)
The Teacher-Student Framework (TSF) is a reinforcement learning setting where a teacher agent guards the training of a student agent by intervening and providing online demonstrations. Assuming optimal, the teacher policy has the perfect timing and capability to intervene in the learning process of the student agent, p...
[ "reinforcement learning", "guarded policy optimization", "imperfect demonstrations", "shared control", "metadrive simulator" ]
null
6,202
2303.01728
title_snapshot
[ -0.0021667731925845146, -0.050078172236680984, -0.0030462078284472227, 0.0662599578499794, 0.03558483347296715, 0.010392095893621445, 0.023585611954331398, 0.0033188702072948217, -0.019290093332529068, -0.022489946335554123, -0.006287708878517151, 0.005431996192783117, -0.061938486993312836,...
Fast Nonlinear Vector Quantile Regression
https://openreview.net/forum?id=UxqUgchwXkK
[ "Aviv A. Rosenberg", "Sanketh Vedula", "Yaniv Romano", "Alexander Bronstein" ]
Poster
Probabilistic Methods (eg, variational inference, causal inference, Gaussian processes)
$$ \newcommand{\rvar}[1]{\mathrm {#1}} \newcommand{\rvec}[1]{\boldsymbol{\mathrm{#1}}} $$ Quantile regression (QR) is a powerful tool for estimating one or more conditional quantiles of a target variable $\rvar{Y}$ given explanatory features $\rvec{X}$. A limitation of QR is that it is only defined for scalar target va...
[ "optimal transport", "quantile regression", "vector quantiles", "uncertainty quantification", "multi-output regression", "conformal prediction", "software" ]
null
6,201
2205.14977
title_snapshot
[ -0.016065077856183052, -0.031083565205335617, 0.01094880048185587, 0.026009507477283478, 0.03726271539926529, 0.07213543355464935, 0.01847892627120018, 0.000027445081286714412, -0.030762091279029846, -0.03791961818933487, -0.0058686635456979275, 0.01838877610862255, -0.06956718862056732, 0...
Leveraging Large Language Models for Multiple Choice Question Answering
https://openreview.net/forum?id=yKbprarjc5B
[ "Joshua Robinson", "David Wingate" ]
Poster
Applications (eg, speech processing, computer vision, NLP)
While large language models (LLMs) like GPT-3 have achieved impressive results on multiple choice question answering (MCQA) tasks in the zero, one, and few-shot settings, they generally lag behind the MCQA state of the art (SOTA). MCQA tasks have traditionally been presented to LLMs like cloze tasks. An LLM is conditio...
[ "NLP", "language models", "multiple choice question answering", "symbol binding", "GPT-3", "Codex" ]
null
6,195
2210.12353
title_snapshot
[ -0.013064540922641754, -0.030637681484222412, -0.013973190449178219, 0.05431386083364487, 0.05737246572971344, 0.03951770067214966, 0.00659413356333971, 0.03209691867232323, -0.026878299191594124, 0.012815442867577076, -0.03017360530793667, 0.06014501675963402, -0.06575264781713486, -0.021...
Learning with Logical Constraints but without Shortcut Satisfaction
https://openreview.net/forum?id=M2unceRvqhh
[ "Zenan Li", "Zehua Liu", "Yuan Yao", "Jingwei Xu", "Taolue Chen", "Xiaoxing Ma", "Jian L\\\"{u}" ]
Notable-top-25%
Deep Learning and representational learning
Recent studies have started to explore the integration of logical knowledge into deep learning via encoding logical constraints as an additional loss function. However, existing approaches tend to vacuously satisfy logical constraints through shortcuts, failing to fully exploit the knowledge. In this paper, we present ...
[ "training with logical constraints", "logical formula encoding", "variational learning", "stochastic gradient descent ascent" ]
null
6,190
2403.00329
title_snapshot
[ -0.0339374765753746, -0.01189972274005413, -0.02754872664809227, 0.06802009046077728, 0.03746217116713524, 0.047627732157707214, 0.021702313795685768, 0.004696030635386705, -0.018657952547073364, -0.019107628613710403, -0.04583502560853958, 0.053756874054670334, -0.05816012993454933, -0.02...
Certified Training: Small Boxes are All You Need
https://openreview.net/forum?id=7oFuxtJtUMH
[ "Mark Niklas Mueller", "Franziska Eckert", "Marc Fischer", "Martin Vechev" ]
Notable-top-25%
Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)
To obtain, deterministic guarantees of adversarial robustness, specialized training methods are used. We propose, SABR, a novel such certified training method, based on the key insight that propagating interval bounds for a small but carefully selected subset of the adversarial input region is sufficient to approximate...
[ "Certified Training", "Certified Robustness", "Adversarial Robustness", "Robustness Verification" ]
null
6,189
2210.04871
title_snapshot
[ -0.010985660366714, -0.025111043825745583, -0.014608296565711498, 0.05718785151839256, 0.02250661887228489, 0.006213861983269453, 0.039370693266391754, -0.04383596032857895, -0.022518496960401535, -0.010799015872180462, -0.017187176272273064, 0.03351179137825966, -0.06158880144357681, -0.0...
Regression with Label Differential Privacy
https://openreview.net/forum?id=h9O0wsmL-cT
[ "Badih Ghazi", "Pritish Kamath", "Ravi Kumar", "Ethan Leeman", "Pasin Manurangsi", "Avinash Varadarajan", "Chiyuan Zhang" ]
Poster
Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)
We study the task of training regression models with the guarantee of _label_ differential privacy (DP). Based on a global prior distribution of 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...
[ "label differential privacy", "regression" ]
null
6,186
2212.06074
title_snapshot
[ 0.01037016324698925, -0.00015779773821122944, -0.011447107419371605, 0.07424631714820862, 0.04508040472865105, 0.04265186935663223, 0.025595106184482574, -0.05684296041727066, -0.020820552483201027, -0.007566187996417284, -0.013237320818006992, 0.004446254577487707, -0.0677511990070343, 0....
Hierarchical Abstraction for Combinatorial Generalization in Object Rearrangement
https://openreview.net/forum?id=fGG6vHp3W9W
[ "Michael Chang", "Alyssa Li Dayan", "Franziska Meier", "Thomas L. Griffiths", "Sergey Levine", "Amy Zhang" ]
Poster
Deep Learning and representational learning
Object rearrangement is a challenge for embodied agents because solving these tasks requires generalizing across a combinatorially large set of configurations of entities and their locations. Worse, the representations of these entities are unknown and must be inferred from sensory percepts. We present a hierarchical a...
[ "objects", "combinatorial generalization", "abstraction", "rearrangement", "slots", "binding", "hierarchy", "compositionality", "symmetry", "independence", "graph search" ]
null
6,184
2303.11373
title_judge
[ -0.030422987416386604, 0.0004931365256197751, -0.0009735973435454071, 0.016447247937321663, 0.03700560703873634, 0.007227152120321989, 0.02348433993756771, 0.0007829554961062968, -0.05452575534582138, -0.045508041977882385, -0.03471921384334564, 0.009270858019590378, -0.06026558205485344, ...
Transfer NAS with Meta-learned Bayesian Surrogates
https://openreview.net/forum?id=paGvsrl4Ntr
[ "Gresa Shala", "Thomas Elsken", "Frank Hutter", "Josif Grabocka" ]
Notable-top-5%
Deep Learning and representational learning
While neural architecture search (NAS) is an intensely-researched area, approaches typically still suffer from either (i) high computational costs or (ii) lack of robustness across datasets and experiments. Furthermore, most methods start searching for an optimal architecture from scratch, ignoring prior knowledge. Thi...
[]
null
6,182
null
null
[ 0.014494898729026318, -0.012814103625714779, -0.011704770848155022, 0.04849889129400253, 0.04733595997095108, 0.046022314578294754, 0.021401025354862213, -0.008473739959299564, 0.006345651112496853, -0.04575972631573677, 0.0041556088253855705, -0.0038029940333217382, -0.035841621458530426, ...
Selective Frequency Network for Image Restoration
https://openreview.net/forum?id=tyZ1ChGZIKO
[ "Yuning Cui", "Yi Tao", "Zhenshan Bing", "Wenqi Ren", "Xinwei Gao", "Xiaochun Cao", "Kai Huang", "Alois Knoll" ]
Poster
Applications (eg, speech processing, computer vision, NLP)
Image restoration aims to reconstruct the latent sharp image from its corrupted counterpart. Besides dealing with this long-standing task in the spatial domain, a few approaches seek solutions in the frequency domain in consideration of the large discrepancy between spectra of sharp/degraded image pairs. However, these...
[ "Image restoration", "Frequency domain", "Frequency selection" ]
null
6,167
null
null
[ 0.02151813730597496, -0.00954914279282093, 0.036307234317064285, 0.015941957011818886, 0.04760035499930382, 0.014931268990039825, 0.007530709262937307, 0.016021497547626495, -0.04632746800780296, -0.07339160889387131, -0.010912247933447361, 0.028661439195275307, -0.03562020882964134, -0.01...
Scaling Up Probabilistic Circuits by Latent Variable Distillation
https://openreview.net/forum?id=067CGykiZTS
[ "Anji Liu", "Honghua Zhang", "Guy Van den Broeck" ]
Notable-top-5%
Probabilistic Methods (eg, variational inference, causal inference, Gaussian processes)
Probabilistic Circuits (PCs) are a unified framework for tractable probabilistic models that support efficient computation of various probabilistic queries (e.g., marginal probabilities). One key challenge is to scale PCs to model large and high-dimensional real-world datasets: we observe that as the number of paramete...
[]
null
6,160
2210.04398
title_snapshot
[ 0.018301082774996758, -0.026100164279341698, -0.021398792043328285, 0.06821666657924652, 0.04786321148276329, 0.015175549313426018, 0.03254028782248497, 0.005652311723679304, -0.025034714490175247, -0.019255368039011955, -0.0031702984124422073, -0.01682678982615471, -0.05029445141553879, 0...
Improving Differentiable Neural Architecture Search by Encouraging Transferability
https://openreview.net/forum?id=Tl8OmiibP99
[ "Parth Sheth", "Pengtao Xie" ]
Poster
Deep Learning and representational learning
Differentiable neural architecture search methods are increasingly popular due to their computational efficiency. However, these methods have unsatisfactory generalizability and stability. Their searched architectures are often degenerate with a dominant number of skip connections and perform unsatisfactorily on...
[]
null
6,154
null
null
[ -0.02885628491640091, -0.013403196819126606, 0.001719258725643158, 0.030252957716584206, 0.05565224960446358, 0.0379980243742466, 0.017062919214367867, -0.020089302211999893, 0.0008281050249934196, -0.018171679228544235, 0.007920104078948498, 0.0025093068834394217, -0.030108420178294182, 0...
MA-BERT: Towards Matrix Arithmetic-only BERT Inference by Eliminating Complex Non-Linear Functions
https://openreview.net/forum?id=HtAfbHa7LAL
[ "Neo Wei Ming", "Zhehui Wang", "Cheng Liu", "Rick Siow Mong Goh", "Tao Luo" ]
Poster
Deep Learning and representational learning
Due to their superior results, Transformer-based models such as BERT have become de facto standards in many Natural Language Processing (NLP) applications. However, the intensive use of complex non-linear functions within the Transformer architecture impairs its computing efficiency and complicates corresponding accele...
[ "BERT", "Efficient inference", "Matrix arithmetic-only", "Eleminate non-linear functions" ]
null
6,153
null
null
[ -0.029754621908068657, -0.002945162123069167, 0.02502545528113842, 0.02827434055507183, 0.01873355731368065, 0.05640615150332451, 0.008587406948208809, -0.010915996506810188, -0.022820619866251945, 0.017077963799238205, -0.005356743931770325, 0.05239631235599518, -0.07403653860092163, 0.01...
Efficient Certified Training and Robustness Verification of Neural ODEs
https://openreview.net/forum?id=KyoVpYvWWnK
[ "Mustafa Zeqiri", "Mark Niklas Mueller", "Marc Fischer", "Martin Vechev" ]
Poster
Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)
Neural Ordinary Differential Equations (NODEs) are a novel neural architecture, built around initial value problems with learned dynamics which are solved during inference. Thought to be inherently more robust against adversarial perturbations, they were recently shown to be vulnerable to strong adversarial attacks, hi...
[ "Neural ODEs", "Adversarial Robustness", "Certified Robustness", "Robustness Verification", "Certified Training" ]
null
6,148
2303.05246
title_snapshot
[ -0.04656210541725159, -0.018150694668293, -0.0067364987917244434, 0.06346215307712555, 0.033108748495578766, 0.03114497847855091, 0.02143058367073536, -0.02914474345743656, -0.021436940878629684, -0.024323979392647743, 0.03429648280143738, -0.0015674799215048552, -0.06376897543668747, 0.00...
Arbitrary Virtual Try-on Network: Characteristics Representation and Trade-off between Body and Clothing
https://openreview.net/forum?id=d8mr8lKIZ3n
[ "Yu Liu", "Mingbo Zhao", "Zhao Zhang", "Jicong Fan", "Yang Lou", "Shuicheng YAN" ]
Poster
Deep Learning and representational learning
Deep learning based virtual try-on system has achieved some encouraging progress recently, but there still remain several big challenges that need to be solved, such as trying on arbitrary clothes of all types, trying on the clothes from one category to another and generating image-realistic results with few artifacts....
[ "Deep Learning", "Virtual Try-on", "Generative Adversarial Networks", "Artificial Intelligence in Fashion" ]
null
6,147
2111.12346
title_judge
[ 0.03259420767426491, -0.021662261337041855, 0.013667359948158264, 0.03694513067603111, 0.041492220014333725, 0.02661329135298729, 0.03956464305520058, 0.008734866976737976, -0.021970953792333603, -0.0682952031493187, -0.038985397666692734, -0.024362212046980858, -0.056729063391685486, -0.0...
UL2: Unifying Language Learning Paradigms
https://openreview.net/forum?id=6ruVLB727MC
[ "Yi Tay", "Mostafa Dehghani", "Vinh Q. Tran", "Xavier Garcia", "Jason Wei", "Xuezhi Wang", "Hyung Won Chung", "Dara Bahri", "Tal Schuster", "Steven Zheng", "Denny Zhou", "Neil Houlsby", "Donald Metzler" ]
Poster
Deep Learning and representational learning
Existing pre-trained models are generally geared towards a particular class of problems. To date, there seems to be still no consensus on what the right architecture and pre-training setup should be. This paper presents a unified framework for pre-training models that are universally effective across datasets and setup...
[ "language models", "pretraining", "transformers" ]
null
6,145
2205.05131
title_snapshot
[ -0.03168230131268501, -0.033620841801166534, 0.008277098648250103, 0.030290665104985237, 0.01897689327597618, 0.0414983294904232, 0.02235276997089386, 0.027930067852139473, -0.0194846298545599, 0.0033965427428483963, -0.04423156753182411, 0.04052527993917465, -0.0688486248254776, -0.024948...
CASR: Generating Complex Sequences with Autoregressive Self-Boost Refinement
https://openreview.net/forum?id=SVl1w1u3InX
[ "Hongwei Han", "Mengyu Zhou", "Shi Han", "Xiu Li", "Dongmei Zhang" ]
Poster
Generative models
There are sequence generation tasks where the best order to generate the target sequence is not left-to-right. For example, an answer to the Sudoku game, a structured code like s-expression, and even a logical natural language answer where the analysis may be generated after the decision. We define the target sequences...
[ "self-boost refinement", "complex answers", "autoregressive generation" ]
null
6,141
null
null
[ -0.011788719333708286, -0.031750015914440155, 0.014275334775447845, 0.03391633182764053, 0.022017579525709152, 0.0387982577085495, 0.021423552185297012, -0.00872948206961155, 0.004468577913939953, -0.03587191551923752, -0.018086154013872147, 0.009343576617538929, -0.03427823260426521, -0.0...
Bitrate-Constrained DRO: Beyond Worst Case Robustness To Unknown Group Shifts
https://openreview.net/forum?id=2QzNuaRHn4Z
[ "Amrith Setlur", "Don Dennis", "Benjamin Eysenbach", "Aditi Raghunathan", "Chelsea Finn", "Virginia Smith", "Sergey Levine" ]
Poster
Deep Learning and representational learning
Training machine learning models robust to distribution shifts is critical for real-world applications. Some robust training algorithms (e.g., Group DRO) specialize to group shifts and require group information on all training points. Other methods (e.g., CVaR DRO) that do not need group annotations can be overly conse...
[ "Robustness", "Distribution shift", "Group Shift" ]
null
6,129
2302.02931
title_snapshot
[ -0.018371514976024628, -0.03193781524896622, 0.02696400322020054, 0.04278840124607086, 0.01348850503563881, 0.004671324975788593, 0.019776567816734314, 0.0006446876213885844, -0.02366585098206997, -0.04437755048274994, -0.05258258804678917, 0.03044796735048294, -0.06091710180044174, 0.0172...
Feature selection and low test error in shallow low-rotation ReLU networks
https://openreview.net/forum?id=swEskiem99
[ "Matus Telgarsky" ]
Poster
Deep Learning and representational learning
This work establishes low test error of gradient flow (GF) and stochastic gradient descent (SGD) on two-layer ReLU networks with standard initialization scale, in three regimes where key sets of weights rotate little (either naturally due to GF and SGD, or due to an artificial constraint), and making use of margins as ...
[ "gradient descent", "gradient flow", "margin maximization", "test error", "neural collapse", "generalization" ]
null
6,124
null
null
[ -0.03555748984217644, -0.029791465029120445, 0.03002401627600193, 0.04524881765246391, 0.031748589128255844, 0.061677832156419754, 0.03244120255112648, 0.00523042818531394, -0.03254009038209915, -0.02260582335293293, 0.019457358866930008, 0.0023348950780928135, -0.06019456684589386, -0.007...
Backpropagation through Combinatorial Algorithms: Identity with Projection Works
https://openreview.net/forum?id=JZMR727O29
[ "Subham Sekhar Sahoo", "Anselm Paulus", "Marin Vlastelica", "Vít Musil", "Volodymyr Kuleshov", "Georg Martius" ]
Poster
Deep Learning and representational learning
Embedding discrete solvers as differentiable layers has given modern deep learning architectures combinatorial expressivity and discrete reasoning capabilities. The derivative of these solvers is zero or undefined, therefore a meaningful replacement is crucial for effective gradient-based learning. Prior works rely on ...
[ "combinatorial optimization", "deep learning", "representation learning", "gradient descent", "backpropagation", "argmin differentiation", "deep graph matching", "retrieval" ]
null
6,123
2205.15213
title_snapshot
[ -0.02917732484638691, -0.0221567265689373, -0.010766655206680298, 0.04397503286600113, 0.03644415736198425, 0.03707615286111832, 0.010295893996953964, -0.028100615367293358, -0.03857841715216637, -0.04391461983323097, 0.013396626338362694, 0.0023207361809909344, -0.03154616802930832, 0.001...
Coupled Multiwavelet Operator Learning for Coupled Differential Equations
https://openreview.net/forum?id=kIo_C6QmMOM
[ "Xiongye Xiao", "Defu Cao", "Ruochen Yang", "Gaurav Gupta", "Gengshuo Liu", "Chenzhong Yin", "Radu Balan", "Paul Bogdan" ]
Poster
Machine Learning for Sciences (eg biology, physics, health sciences, social sciences, climate/sustainability )
Coupled partial differential equations (PDEs) are key tasks in modeling the complex dynamics of many physical processes. Recently, neural operators have shown the ability to solve PDEs by learning the integral kernel directly in Fourier/Wavelet space, so the difficulty of solving the coupled PDEs depends on dealing wit...
[ "Neural operators", "coupled differential equations", "multiwavelet transform", "partial differential equations" ]
null
6,120
2303.02304
title_judge
[ -0.059519775211811066, -0.030756179243326187, 0.0470518097281456, 0.038589488714933395, 0.03536755591630936, 0.025708099827170372, -0.0019613897893577814, -0.003327305195853114, -0.032754722982645035, -0.0756063163280487, 0.008656038902699947, 0.023983893916010857, -0.04130750149488449, 0....
Mid-Vision Feedback
https://openreview.net/forum?id=4oLK1_k71Tz
[ "Michael Maynord", "Eadom T Dessalene", "Cornelia Fermuller", "Yiannis Aloimonos" ]
Poster
Applications (eg, speech processing, computer vision, NLP)
Feedback plays a prominent role in biological vision, where perception is modulated based on agents' evolving expectations and world model. We introduce a novel mechanism which modulates perception based on high level categorical expectations: Mid-Vision Feedback (MVF). MVF associates high level contexts with linear tr...
[]
null
6,118
null
null
[ 0.0055127800442278385, 0.020332444459199905, -0.004676642827689648, 0.018712157383561134, 0.027378024533391, -0.0032944295089691877, 0.031270913779735565, 0.02569994516670704, -0.0442662239074707, -0.047369811683893204, -0.02129920944571495, 0.0012381519190967083, -0.08923723548650742, -0....
Safe Reinforcement Learning From Pixels Using a Stochastic Latent Representation
https://openreview.net/forum?id=b39dQt_uffW
[ "Yannick Hogewind", "Thiago D. Simão", "Tal Kachman", "Nils Jansen" ]
Poster
Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)
We address the problem of safe reinforcement learning from pixel observations. Inherent challenges in such settings are (1) a trade-off between reward optimization and adhering to safety constraints, (2) partial observability, and (3) high-dimensional observations. We formalize the problem in a constrained, partially o...
[ "safety", "reinforcement learning", "safe reinforcement learning", "constrained Markov decision process", "partially observable Markov decision process", "MDP", "POMDP" ]
null
6,112
2210.01801
title_snapshot
[ -0.01831800490617752, -0.0226189773529768, -0.027669735252857208, 0.05802982673048973, 0.02727602794766426, 0.02150909975171089, 0.015170246362686157, 0.013229256495833397, -0.029924316331744194, -0.03945770859718323, -0.027996355667710304, -0.004462997894734144, -0.07695427536964417, -0.0...
TrojText: Test-time Invisible Textual Trojan Insertion
https://openreview.net/forum?id=ja4Lpp5mqc2
[ "Qian Lou", "Yepeng Liu", "Bo Feng" ]
Poster
Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)
In Natural Language Processing (NLP), intelligent neuron models can be susceptible to textual Trojan attacks. Such attacks occur when Trojan models behave normally for standard inputs but generate malicious output for inputs that contain a specific trigger. Syntactic-structure triggers, which are invisible, are becomin...
[ "Textual", "Trojan", "Backdoor", "Syntactic", "Trigger", "Invisible", "Attack", "Defense", "Test-time" ]
null
6,102
2303.02242
title_snapshot
[ -0.008654310368001461, -0.02695976197719574, -0.03829284757375717, 0.05010590702295303, 0.03159847483038902, 0.0032887135166674852, 0.050257567316293716, 0.022820986807346344, -0.025809185579419136, -0.008555361069738865, -0.02414657734334469, 0.026683295145630836, -0.046551138162612915, -...
Multi-Objective Online Learning
https://openreview.net/forum?id=dKkMnCWfVmm
[ "Jiyan Jiang", "Wenpeng Zhang", "Shiji Zhou", "Lihong Gu", "Xiaodong Zeng", "Wenwu Zhu" ]
Notable-top-25%
Optimization (eg, convex and non-convex optimization)
This paper presents a systematic study of multi-objective online learning. We first formulate the framework of Multi-Objective Online Convex Optimization, which encompasses a novel multi-objective regret. This regret is built upon a sequence-wise extension of the commonly used discrepancy metric Pareto suboptimality ga...
[]
null
6,099
null
null
[ -0.04723668098449707, 0.002806968754157424, 0.013586994260549545, 0.041709791868925095, 0.03151801601052284, 0.04334636032581329, 0.01616670936346054, 0.0325871966779232, -0.027899835258722305, -0.04753904417157173, 0.00014105472655501217, 0.011011316440999508, -0.058973319828510284, -0.04...
Improved Training of Physics-Informed Neural Networks Using Energy-Based Priors: a Study on Electrical Impedance Tomography
https://openreview.net/forum?id=zqkfJA6R1-r
[ "Akarsh Pokkunuru", "Pedram Rooshenas", "Thilo Strauss", "Anuj Abhishek", "Taufiquar Khan" ]
Poster
Machine Learning for Sciences (eg biology, physics, health sciences, social sciences, climate/sustainability )
Physics-informed neural networks (PINNs) are attracting significant attention for solving partial differential equation (PDE) based inverse problems, including electrical impedance tomography (EIT). EIT is non-linear and especially its inverse problem is highly ill-posed. Therefore, successful training of PINN is extre...
[ "Physics-informed neural networks", "electrical impedance tomography", "energy-based models" ]
null
6,098
null
null
[ -0.024464810267090797, -0.02242588810622692, -0.012115059420466423, 0.03819656744599342, 0.0245275367051363, 0.013953019864857197, 0.010470299050211906, -0.030704161152243614, -0.04673996567726135, -0.051473360508680344, -0.017642982304096222, 0.006112329196184874, -0.034908413887023926, -...
A Kernel Perspective of Skip Connections in Convolutional Networks
https://openreview.net/forum?id=6H_uOfcwiVh
[ "Daniel Barzilai", "Amnon Geifman", "Meirav Galun", "Ronen Basri" ]
Notable-top-5%
Theory (eg, control theory, learning theory, algorithmic game theory)
Over-parameterized residual networks (ResNets) are amongst the most successful convolutional neural architectures for image processing. Here we study their properties through their Gaussian Process and Neural Tangent kernels. We derive explicit formulas for these kernels, analyze their spectra, and provide bounds on th...
[]
null
6,089
2211.14810
title_snapshot
[ -0.019331738352775574, -0.009446482174098492, 0.010942568071186543, 0.07129917293787003, 0.04235469922423363, 0.03842093423008919, 0.009642032906413078, 0.019118621945381165, -0.014933179132640362, -0.055546291172504425, -0.0014942496782168746, 0.023636259138584137, -0.048483528196811676, ...
Ordered GNN: Ordering Message Passing to Deal with Heterophily and Over-smoothing
https://openreview.net/forum?id=wKPmPBHSnT6
[ "Yunchong Song", "Chenghu Zhou", "Xinbing Wang", "Zhouhan Lin" ]
Poster
Deep Learning and representational learning
Most graph neural networks follow the message passing mechanism. However, it faces the over-smoothing problem when multiple times of message passing is applied to a graph, causing indistinguishable node representations and prevents the model to effectively learn dependencies between farther-away nodes. On the other han...
[ "GNN", "heterophily", "over-smoothing" ]
null
6,087
2302.01524
title_snapshot
[ -0.04207433760166168, -0.019447989761829376, -0.009053146466612816, 0.0527673177421093, 0.024564480409026146, 0.02615133486688137, 0.02467542700469494, 0.030796142295002937, -0.031238066032528877, -0.04600563645362854, 0.020409036427736282, -0.0398399792611599, -0.08619707077741623, 0.0138...
Sparse Distributed Memory is a Continual Learner
https://openreview.net/forum?id=JknGeelZJpHP
[ "Trenton Bricken", "Xander Davies", "Deepak Singh", "Dmitry Krotov", "Gabriel Kreiman" ]
Poster
Neuroscience and Cognitive Science (e.g., neural coding, brain-computer interfaces)
Continual learning is a problem for artificial neural networks that their biological counterparts are adept at solving. Building on work using Sparse Distributed Memory (SDM) to connect a core neural circuit with the powerful Transformer model, we create a modified Multi-Layered Perceptron (MLP) that is a strong contin...
[ "Sparse Distributed Memory", "Sparsity", "Top-K Activation", "Continual Learning", "Biologically Inspired" ]
null
6,086
2303.11934
title_snapshot
[ -0.0019244890427216887, -0.025120528414845467, -0.039820458739995956, 0.02379739284515381, 0.03459590673446655, 0.02270837128162384, 0.029655713587999344, 0.02843432128429413, -0.06207654997706413, -0.023654755204916, 0.039058562368154526, 0.0010081608779728413, -0.06410536915063858, 0.021...
FLIP: A Provable Defense Framework for Backdoor Mitigation in Federated Learning
https://openreview.net/forum?id=Xo2E217_M4n
[ "Kaiyuan Zhang", "Guanhong Tao", "Qiuling Xu", "Siyuan Cheng", "Shengwei An", "Yingqi Liu", "Shiwei Feng", "Guangyu Shen", "Pin-Yu Chen", "Shiqing Ma", "Xiangyu Zhang" ]
Poster
General Machine Learning (ie none of the above)
Federated Learning (FL) is a distributed learning paradigm that enables different parties to train a model together for high quality and strong privacy protection. In this scenario, individual participants may get compromised and perform backdoor attacks by poisoning the data (or gradients). Existing work on robust agg...
[ "Federated learning", "backdoor mitigation" ]
null
6,082
2210.12873
title_snapshot
[ -0.013128980994224548, -0.04456755518913269, 0.0034066319931298494, 0.044217299669981, 0.053110767155885696, -0.016975078731775284, 0.05309632793068886, -0.015428158454596996, -0.004820206668227911, -0.02339804172515869, 0.02113998867571354, -0.0023200830910354853, -0.0344759076833725, -0....
UniMax: Fairer and More Effective Language Sampling for Large-Scale Multilingual Pretraining
https://openreview.net/forum?id=kXwdL1cWOAi
[ "Hyung Won Chung", "Xavier Garcia", "Adam Roberts", "Yi Tay", "Orhan Firat", "Sharan Narang", "Noah Constant" ]
Poster
Deep Learning and representational learning
Pretrained multilingual large language models have typically used heuristic temperature-based sampling to balance between different languages. However previous work has not systematically evaluated the efficacy of different pretraining language distributions across model scales. In this paper, we propose a new sampling...
[ "Keywords: multilingual", "pretraining", "language models", "language sampling", "language distribution", "low-resource languages", "overfitting" ]
null
6,072
2304.09151
title_snapshot
[ -0.0013562749372795224, -0.044198423624038696, -0.007672359235584736, 0.044170401990413666, 0.022571509703993797, 0.02744496613740921, 0.021871419623494148, 0.03654065728187561, -0.027570517733693123, -0.026878856122493744, -0.009466397576034069, 0.02503044717013836, -0.07564249634742737, ...
GNNInterpreter: A Probabilistic Generative Model-Level Explanation for Graph Neural Networks
https://openreview.net/forum?id=rqq6Dh8t4d
[ "Xiaoqi Wang", "Han Wei Shen" ]
Poster
Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)
Recently, Graph Neural Networks (GNNs) have significantly advanced the performance of machine learning tasks on graphs. However, this technological breakthrough makes people wonder: how does a GNN make such decisions, and can we trust its prediction with high confidence? When it comes to some critical fields, such as b...
[ "AI Interpretability", "Graph Neural Networks", "Model-Level Explanation of Neural Networks" ]
null
6,066
2209.07924
title_snapshot
[ -0.02739037200808525, -0.03621144965291023, -0.011981087736785412, 0.0793810710310936, 0.031916871666908264, 0.01145808957517147, 0.013464823365211487, -0.0005677869776263833, -0.00396001385524869, -0.04553043469786644, -0.011747372336685658, 0.009310931898653507, -0.04465016350150108, 0.0...
Rethinking Symbolic Regression: Morphology and Adaptability in the Context of Evolutionary Algorithms
https://openreview.net/forum?id=OPGy07PojsZ
[ "Kei Sen Fong", "Shelvia Wongso", "Mehul Motani" ]
Poster
General Machine Learning (ie none of the above)
Symbolic Regression (SR) is the well-studied problem of finding closed-form analytical expressions that describe the relationship between variables in a measurement dataset. In this paper, we rethink SR from two perspectives: morphology and adaptability. Morphology: Current SR algorithms typically use several man-made ...
[]
null
6,064
null
null
[ -0.035992562770843506, -0.008849493227899075, -0.01802639104425907, 0.027859892696142197, 0.06145385280251503, 0.0729767233133316, 0.0482499822974205, -0.024215713143348694, -0.03946268931031227, -0.04555215686559677, -0.010196500457823277, 0.04194745048880577, -0.06663773953914642, 0.0071...
On Pre-training Language Model for Antibody
https://openreview.net/forum?id=zaq4LV55xHl
[ "Danqing Wang", "Fei YE", "Hao Zhou" ]
Poster
Machine Learning for Sciences (eg biology, physics, health sciences, social sciences, climate/sustainability )
Antibodies are vital proteins offering robust protection for the human body from pathogens. The development of general protein and antibody-specific pre-trained language models both facilitate antibody prediction tasks. However, there have been limited studies that comprehensively explore the representation capability ...
[]
null
6,062
null
null
[ -0.01136567909270525, -0.0008342319633811712, -0.009464563801884651, 0.02371177263557911, 0.03182518482208252, 0.011984203942120075, 0.05867667868733406, 0.02057909034192562, 0.0023726425133645535, 0.0014359875349327922, -0.015400765463709831, 0.02698494680225849, -0.08963925391435623, 0.0...
Learning to reason over visual objects
https://openreview.net/forum?id=uR6x8Be7o_M
[ "Shanka Subhra Mondal", "Taylor Whittington Webb", "Jonathan Cohen" ]
Poster
Deep Learning and representational learning
A core component of human intelligence is the ability to identify abstract patterns inherent in complex, high-dimensional perceptual data, as exemplified by visual reasoning tasks such as Raven’s Progressive Matrices (RPM). Motivated by the goal of designing AI systems with this capacity, recent work has focused on eva...
[]
null
6,060
2303.02260
title_snapshot
[ -0.010923475958406925, 0.011688495054841042, 0.018697887659072876, 0.05167971923947334, 0.018407106399536133, -0.0003507500805426389, 0.029451297596096992, 0.018285874277353287, -0.05045085400342941, -0.017059605568647385, -0.02057436853647232, 0.041562341153621674, -0.0650152713060379, 0....
Imitating Graph-Based Planning with Goal-Conditioned Policies
https://openreview.net/forum?id=6lUEy1J5R7p
[ "Junsu Kim", "Younggyo Seo", "Sungsoo Ahn", "Kyunghwan Son", "Jinwoo Shin" ]
Poster
Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)
Recently, graph-based planning algorithms have gained much attention to solve goal-conditioned reinforcement learning (RL) tasks: they provide a sequence of subgoals to reach the target-goal, and the agents learn to execute subgoal-conditioned policies. However, the sample-efficiency of such RL schemes still remains a ...
[ "Reinforcement Learning", "Goal-Conditioned Reinforcement Learning" ]
null
6,058
2303.11166
title_snapshot
[ -0.013333809562027454, -0.0488811656832695, -0.008005975745618343, 0.035508375614881516, 0.05011282116174698, 0.01946805790066719, 0.024401895701885223, -0.01728312112390995, -0.01857241429388523, -0.03916772082448006, -0.01286829449236393, -0.0023833196610212326, -0.0724748745560646, -0.0...
Seeing Differently, Acting Similarly: Heterogeneously Observable Imitation Learning
https://openreview.net/forum?id=3ULaIHxn9u7
[ "Xin-Qiang Cai", "Yao-Xiang Ding", "Zixuan Chen", "Yuan Jiang", "Masashi Sugiyama", "Zhi-Hua Zhou" ]
Notable-top-25%
Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)
In many real-world imitation learning tasks, the demonstrator and the learner have to act under different observation spaces. This situation brings significant obstacles to existing imitation learning approaches, since most of them learn policies under homogeneous observation spaces. On the other hand, previous studies...
[ "Imitation Learning", "Heterogeneous Observation", "Importance Weighting", "Learning with Rejection" ]
null
6,056
2106.09256
title_snapshot
[ -0.018952803686261177, -0.009577608667314053, -0.011312293820083141, 0.01436128281056881, 0.022656839340925217, 0.03540617600083351, 0.03174032270908356, 0.0215822272002697, -0.0494219996035099, -0.03233417496085167, -0.004701489582657814, -0.007127232849597931, -0.09591776132583618, -0.01...
A theoretical study of inductive biases in contrastive learning
https://openreview.net/forum?id=AuEgNlEAmed
[ "Jeff Z. HaoChen", "Tengyu Ma" ]
Poster
Theory (eg, control theory, learning theory, algorithmic game theory)
Understanding self-supervised learning is important but challenging. Previous theoretical works study the role of pretraining losses, and view neural networks as general black boxes. However, the recent work of [Saunshi et al.] argues that the model architecture --- a component largely ignored by previous works --- als...
[ "theory of self-supervised learning", "theory of contrastive learning" ]
null
6,054
2211.14699
title_snapshot
[ -0.01525939255952835, -0.01962238922715187, -0.01721438206732273, 0.061906445771455765, 0.015148263424634933, 0.024128306657075882, 0.021812649443745613, 0.019643424078822136, -0.038030121475458145, -0.01277482882142067, -0.008442483842372894, 0.015545131638646126, -0.055547114461660385, 0...
Combinatorial Pure Exploration of Causal Bandits
https://openreview.net/forum?id=pBBsrPzq7aF
[ "Nuoya Xiong", "Wei Chen" ]
Poster
Theory (eg, control theory, learning theory, algorithmic game theory)
The combinatorial pure exploration of causal bandits is the following online learning task: given a causal graph with unknown causal inference distributions, in each round we choose a subset of variables to intervene or do no intervention, and observe the random outcomes of all random variables, with the goal that usin...
[ "Bandit", "causal bandit", "pure exploration" ]
null
6,052
2206.07883
title_snapshot
[ -0.011237011291086674, -0.024702295660972595, -0.0075622196309268475, 0.04818578064441681, 0.03518921881914139, 0.025003166869282722, 0.04124036058783531, 0.013642491772770882, -0.024234632030129433, -0.0361906923353672, -0.015735670924186707, 0.022548114880919456, -0.04922899603843689, -0...
Computational Language Acquisition with Theory of Mind
https://openreview.net/forum?id=C2ulri4duIs
[ "Andy Liu", "Hao Zhu", "Emmy Liu", "Yonatan Bisk", "Graham Neubig" ]
Poster
Neuroscience and Cognitive Science (e.g., neural coding, brain-computer interfaces)
Unlike current state-of-the-art language models, young children actively acquire language through interactions with their surrounding environment and caretakers. One mechanism that has been argued to be critical to language learning is the ability to infer the mental states of other agents in social environments, coine...
[ "language acquisition", "theory of mind", "referential games", "natural language processing" ]
null
6,049
2303.01502
title_snapshot
[ -0.026086125522851944, 0.01735120825469494, -0.023168183863162994, 0.016683032736182213, 0.03973696008324623, 0.023107634857296944, 0.04815066233277321, 0.0462891161441803, -0.029687395319342613, -0.009976241737604141, -0.051645394414663315, 0.03605847433209419, -0.04247014597058296, -0.00...
A Closer Look at Model Adaptation using Feature Distortion and Simplicity Bias
https://openreview.net/forum?id=wkg_b4-IwTZ
[ "Puja Trivedi", "Danai Koutra", "Jayaraman J. Thiagarajan" ]
Notable-top-25%
Deep Learning and representational learning
Advances in the expressivity of pretrained models have increased interest in the design of adaptation protocols which enable safe and effective transfer learning. Going beyond conventional linear probing (LP) and fine tuning (FT) strategies, protocols that can effectively control feature distortion, i.e., the failure t...
[ "Transfer Learning", "Robustness", "Adaptation", "Data Augmentation" ]
null
6,041
2303.13500
title_snapshot
[ -0.007629928179085255, -0.028650684282183647, -0.018667178228497505, 0.03860783576965332, 0.04553554952144623, 0.005125302821397781, 0.041282910853624344, -0.015030440874397755, -0.004743731580674648, -0.0464433915913105, 0.0018413369543850422, 0.01715298555791378, -0.08007941395044327, -0...
Pareto Invariant Risk Minimization: Towards Mitigating the Optimization Dilemma in Out-of-Distribution Generalization
https://openreview.net/forum?id=esFxSb_0pSL
[ "Yongqiang Chen", "Kaiwen Zhou", "Yatao Bian", "Binghui Xie", "Bingzhe Wu", "Yonggang Zhang", "MA KAILI", "Han Yang", "Peilin Zhao", "Bo Han", "James Cheng" ]
Poster
Deep Learning and representational learning
Recently, there has been a growing surge of interest in enabling machine learning systems to generalize well to Out-of-Distribution (OOD) data. Most efforts are devoted to advancing optimization objectives that regularize models to capture the underlying invariance; however, there often are compromises in the optimizat...
[ "Out-of-Distribution Generalization", "Optimization", "Multi-Objective Optimization", "Causal Invariance" ]
null
6,040
2206.07766
title_snapshot
[ -0.024891411885619164, -0.0021268953569233418, 0.00010276355169480667, 0.0203568022698164, 0.03299662843346596, 0.04208134859800339, 0.0224795863032341, -0.014452368952333927, -0.03780461847782135, -0.018162142485380173, -0.036488212645053864, 0.00046289857709780335, -0.07409477978944778, ...
Understanding and Adopting Rational Behavior by Bellman Score Estimation
https://openreview.net/forum?id=WzGdBqcBicl
[ "Kuno Kim", "Stefano Ermon" ]
Notable-top-25%
Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)
We are interested in solving a class of problems that seek to understand and adopt rational behavior from demonstrations. We may broadly classify these problems into four categories of reward identification, counterfactual analysis, behavior imitation, and behavior transfer. In this work, we make a key observation that...
[ "Inverse Reinforcement Learning" ]
null
6,033
null
null
[ -0.02938169427216053, 0.001385210664011538, -0.009638192132115364, 0.03253760561347008, 0.06254193931818008, 0.027395427227020264, 0.040327947586774826, -0.0007117103668861091, -0.03417274355888367, -0.04154067113995552, 0.0016039613401517272, 0.02863473817706108, -0.07457441836595535, -0....
What Makes Convolutional Models Great on Long Sequence Modeling?
https://openreview.net/forum?id=TGJSPbRpJX-
[ "Yuhong Li", "Tianle Cai", "Yi Zhang", "Deming Chen", "Debadeepta Dey" ]
Poster
Deep Learning and representational learning
Convolutional models have been widely used in multiple domains. However, most existing models only use local convolution, making the model unable to handle long-range dependencies efficiently. Attention overcomes this problem by aggregating global information based on the pair-wise attention score but also makes the co...
[ "Convolutional Neural Network", "Deep Learning Architectures", "Long-range dependence", "Reparameterization" ]
null
6,028
2210.09298
title_snapshot
[ -0.004536824766546488, -0.04472093656659126, 0.006484834011644125, 0.03514996916055679, 0.019037803635001183, 0.03815329074859619, 0.022879749536514282, 0.032067231833934784, -0.0003703880647663027, -0.043296582996845245, 0.0046529811806976795, -0.0004336825804784894, -0.07114493101835251, ...
Editing models with task arithmetic
https://openreview.net/forum?id=6t0Kwf8-jrj
[ "Gabriel Ilharco", "Marco Tulio Ribeiro", "Mitchell Wortsman", "Ludwig Schmidt", "Hannaneh Hajishirzi", "Ali Farhadi" ]
Poster
Deep Learning and representational learning
Changing how pre-trained models behave---e.g., improving their performance on a downstream task or mitigating biases learned during pre-training---is a common practice when developing machine learning systems. In this work, we propose a new paradigm for steering the behavior of neural networks, centered around task vec...
[ "pre-trained models", "model editing", "model patching", "fine-tuning", "transfer learning", "weight interpolation", "merging models" ]
null
6,019
2212.04089
title_snapshot
[ -0.006670802365988493, 0.0004153345653321594, -0.024092070758342743, 0.022490156814455986, 0.05200624838471413, 0.017817921936511993, 0.02517089806497097, 0.014283029362559319, -0.042498938739299774, -0.04556218907237053, -0.018554996699094772, 0.0317830964922905, -0.06391964852809906, -0....