paper_id
string
title
string
paper_url
string
pdf_url
string
authors
list
abstract
large_string
track
string
primary_area
string
doi
string
volume
string
issue
string
pages
string
abstract_source
string
arxiv_id
string
arxiv_id_source
string
10.1609/aaai.v35i8.16849
Provably Secure Federated Learning against Malicious Clients
https://ojs.aaai.org/index.php/AAAI/article/view/16849
https://ojs.aaai.org/index.php/AAAI/article/download/16849/16656
[ "Xiaoyu Cao", "Jinyuan Jia", "Neil Zhenqiang Gong" ]
Federated learning enables clients to collaboratively learn a shared global model without sharing their local training data with a cloud server. However, malicious clients can corrupt the global model to predict incorrect labels for testing examples. Existing defenses against malicious clients leverage Byzantine-robust...
main
Machine Learning
10.1609/aaai.v35i8.16849
35
8
6885-6893
official
2102.01854
title_snapshot
10.1609/aaai.v35i8.16850
Dual Quaternion Knowledge Graph Embeddings
https://ojs.aaai.org/index.php/AAAI/article/view/16850
https://ojs.aaai.org/index.php/AAAI/article/download/16850/16657
[ "Zongsheng Cao", "Qianqian Xu", "Zhiyong Yang", "Xiaochun Cao", "Qingming Huang" ]
In this paper, we study the problem of learning representations of entities and relations in the knowledge graph for the link prediction task. Our idea is based on the observation that the vast majority of the related work only models the relation as a single geometric operation such as translation or rotation, which l...
main
Machine Learning
10.1609/aaai.v35i8.16850
35
8
6894-6902
official
null
null
10.1609/aaai.v35i8.16851
Counterfactual Explanations for Oblique Decision Trees:Exact, Efficient Algorithms
https://ojs.aaai.org/index.php/AAAI/article/view/16851
https://ojs.aaai.org/index.php/AAAI/article/download/16851/16658
[ "Miguel Á. Carreira-Perpiñán", "Suryabhan Singh Hada" ]
We consider counterfactual explanations, the problem of minimally adjusting features in a source input instance so that it is classified as a target class under a given classifier. This has become a topic of recent interest as a way to query a trained model and suggest possible actions to overturn its decision. Mathema...
main
Machine Learning
10.1609/aaai.v35i8.16851
35
8
6903-6911
official
2103.01096
title_snapshot
10.1609/aaai.v35i8.16852
Curriculum Labeling: Revisiting Pseudo-Labeling for Semi-Supervised Learning
https://ojs.aaai.org/index.php/AAAI/article/view/16852
https://ojs.aaai.org/index.php/AAAI/article/download/16852/16659
[ "Paola Cascante-Bonilla", "Fuwen Tan", "Yanjun Qi", "Vicente Ordonez" ]
In this paper we revisit the idea of pseudo-labeling in the context of semi-supervised learning where a learning algorithm has access to a small set of labeled samples and a large set of unlabeled samples. Pseudo-labeling works by applying pseudo-labels to samples in the unlabeled set by using a model trained on combin...
main
Machine Learning
10.1609/aaai.v35i8.16852
35
8
6912-6920
official
2001.06001
title_snapshot
10.1609/aaai.v35i8.16853
Frivolous Units: Wider Networks Are Not Really That Wide
https://ojs.aaai.org/index.php/AAAI/article/view/16853
https://ojs.aaai.org/index.php/AAAI/article/download/16853/16660
[ "Stephen Casper", "Xavier Boix", "Vanessa D'Amario", "Ling Guo", "Martin Schrimpf", "Kasper Vinken", "Gabriel Kreiman" ]
A remarkable characteristic of overparameterized deep neural networks (DNNs) is that their accuracy does not degrade when the network width is increased. Recent evidence suggests that developing compressible representations allows the complexity of large networks to be adjusted for the learning task at hand. However, t...
main
Machine Learning
10.1609/aaai.v35i8.16853
35
8
6921-6929
official
1912.04783
title_snapshot
10.1609/aaai.v35i8.16854
Automated Clustering of High-dimensional Data with a Feature Weighted Mean Shift Algorithm
https://ojs.aaai.org/index.php/AAAI/article/view/16854
https://ojs.aaai.org/index.php/AAAI/article/download/16854/16661
[ "Saptarshi Chakraborty", "Debolina Paul", "Swagatam Das" ]
Mean shift is a simple interactive procedure that gradually shifts data points towards the mode which denotes the highest density of data points in the region. Mean shift algorithms have been effectively used for data denoising, mode seeking, and finding the number of clusters in a dataset in an automated fashion. Howe...
main
Machine Learning
10.1609/aaai.v35i8.16854
35
8
6930-6938
official
2012.10929
title_snapshot
10.1609/aaai.v35i8.16855
High-Confidence Off-Policy (or Counterfactual) Variance Estimation
https://ojs.aaai.org/index.php/AAAI/article/view/16855
https://ojs.aaai.org/index.php/AAAI/article/download/16855/16662
[ "Yash Chandak", "Shiv Shankar", "Philip S. Thomas" ]
Many sequential decision-making systems leverage data collected using prior policies to propose a new policy. For critical applications, it is important that high-confidence guarantees on the new policy’s behavior are provided before deployment, to ensure that the policy will behave as desired. Prior works have studied...
main
Machine Learning
10.1609/aaai.v35i8.16855
35
8
6939-6947
official
2101.09847
title_snapshot
10.1609/aaai.v35i8.16856
A Multi-step-ahead Markov Conditional Forward Model with Cube Perturbations for Extreme Weather Forecasting
https://ojs.aaai.org/index.php/AAAI/article/view/16856
https://ojs.aaai.org/index.php/AAAI/article/download/16856/16663
[ "Chia-Yuan Chang", "Cheng-Wei Lu", "Chuan-Ju Wang" ]
Predicting extreme weather events such as tropical and extratropical cyclones is of vital scientific and societal importance. Of late, machine learning methods have found their way to weather analysis and prediction, but mostly, these methods use machine learning merely as a complement to traditional numerical weather ...
main
Machine Learning
10.1609/aaai.v35i8.16856
35
8
6948-6955
official
null
null
10.1609/aaai.v35i8.16857
Extending Multi-Sense Word Embedding to Phrases and Sentences for Unsupervised Semantic Applications
https://ojs.aaai.org/index.php/AAAI/article/view/16857
https://ojs.aaai.org/index.php/AAAI/article/download/16857/16664
[ "Haw-Shiuan Chang", "Amol Agrawal", "Andrew McCallum" ]
Most unsupervised NLP models represent each word with a single point or single region in semantic space, while the existing multi-sense word embeddings cannot represent longer word sequences like phrases or sentences. We propose a novel embedding method for a text sequence (a phrase or a sentence) where each sequence i...
main
Machine Learning
10.1609/aaai.v35i8.16857
35
8
6956-6965
official
2103.15330
title_snapshot
10.1609/aaai.v35i8.16858
On Online Optimization: Dynamic Regret Analysis of Strongly Convex and Smooth Problems
https://ojs.aaai.org/index.php/AAAI/article/view/16858
https://ojs.aaai.org/index.php/AAAI/article/download/16858/16665
[ "Ting-Jui Chang", "Shahin Shahrampour" ]
The regret bound of dynamic online learning algorithms is often expressed in terms of the variation in the function sequence (V_T) and/or the path-length of the minimizer sequence after T rounds. For strongly convex and smooth functions, Zhang et al. (2017) establish the squared path-length of the minimizer sequence (C...
main
Machine Learning
10.1609/aaai.v35i8.16858
35
8
6966-6973
official
2006.03912
title_judge
10.1609/aaai.v35i8.16859
Provable Benefits of Overparameterization in Model Compression: From Double Descent to Pruning Neural Networks
https://ojs.aaai.org/index.php/AAAI/article/view/16859
https://ojs.aaai.org/index.php/AAAI/article/download/16859/16666
[ "Xiangyu Chang", "Yingcong Li", "Samet Oymak", "Christos Thrampoulidis" ]
Deep networks are typically trained with many more parameters than the size of the training dataset. Recent empirical evidence indicates that the practice of overparameterization not only benefits training large models, but also assists – perhaps counterintuitively – building lightweight models. Specifically, it sugges...
main
Machine Learning
10.1609/aaai.v35i8.16859
35
8
6974-6983
official
2012.08749
title_snapshot
10.1609/aaai.v35i8.16860
Differentially Private Decomposable Submodular Maximization
https://ojs.aaai.org/index.php/AAAI/article/view/16860
https://ojs.aaai.org/index.php/AAAI/article/download/16860/16667
[ "Anamay Chaturvedi", "Huy Lê Nguyễn", "Lydia Zakynthinou" ]
We study the problem of differentially private constrained maximization of decomposable submodular functions. A submodular function is decomposable if it takes the form of a sum of submodular functions. The special case of maximizing a monotone, decomposable submodular function under cardinality constraints is known as...
main
Machine Learning
10.1609/aaai.v35i8.16860
35
8
6984-6992
official
2005.14717
title_snapshot
10.1609/aaai.v35i8.16861
Using Hindsight to Anchor Past Knowledge in Continual Learning
https://ojs.aaai.org/index.php/AAAI/article/view/16861
https://ojs.aaai.org/index.php/AAAI/article/download/16861/16668
[ "Arslan Chaudhry", "Albert Gordo", "Puneet Dokania", "Philip Torr", "David Lopez-Paz" ]
In continual learning, the learner faces a stream of data whose distribution changes over time. Modern neural networks are known to suffer under this setting, as they quickly forget previously acquired knowledge. To address such catastrophic forgetting, many continual learning methods implement different types of exper...
main
Machine Learning
10.1609/aaai.v35i8.16861
35
8
6993-7001
official
2002.08165
title_snapshot
10.1609/aaai.v35i8.16862
Deep Verifier Networks: Verification of Deep Discriminative Models with Deep Generative Models
https://ojs.aaai.org/index.php/AAAI/article/view/16862
https://ojs.aaai.org/index.php/AAAI/article/download/16862/16669
[ "Tong Che", "Xiaofeng Liu", "Site Li", "Yubin Ge", "Ruixiang Zhang", "Caiming Xiong", "Yoshua Bengio" ]
AI Safety is a major concern in many deep learning applications such as autonomous driving. Given a trained deep learning model, an important natural problem is how to reliably verify the model's prediction. In this paper, we propose a novel framework --- deep verifier networks (DVN) to detect unreliable inputs or pred...
main
Machine Learning
10.1609/aaai.v35i8.16862
35
8
7002-7010
official
1911.07421
title_snapshot
10.1609/aaai.v35i8.16823
An Enhanced Advising Model in Teacher-Student Framework using State Categorization
https://ojs.aaai.org/index.php/AAAI/article/view/16823
https://ojs.aaai.org/index.php/AAAI/article/download/16823/16630
[ "Daksh Anand", "Vaibhav Gupta", "Praveen Paruchuri", "Balaraman Ravindran" ]
The teacher-student framework aims to improve the sample efficiency of RL algorithms by deploying an advising mechanism in which a teacher helps a student by guiding its exploration. Prior work in this field has considered an advising mechanism where the teacher advises the student about the optimal action to take in a...
main
Machine Learning
10.1609/aaai.v35i8.16823
35
8
6653-6660
official
null
null
10.1609/aaai.v35i8.16824
On Lipschitz Regularization of Convolutional Layers using Toeplitz Matrix Theory
https://ojs.aaai.org/index.php/AAAI/article/view/16824
https://ojs.aaai.org/index.php/AAAI/article/download/16824/16631
[ "Alexandre Araujo", "Benjamin Negrevergne", "Yann Chevaleyre", "Jamal Atif" ]
This paper tackles the problem of Lipschitz regularization of Convolutional Neural Networks. Lipschitz regularity is now established as a key property of modern deep learning with implications in training stability, generalization, robustness against adversarial examples, etc. However, computing the exact value of the ...
main
Machine Learning
10.1609/aaai.v35i8.16824
35
8
6661-6669
official
2006.08391
title_snapshot
10.1609/aaai.v35i8.16825
The Tractability of SHAP-Score-Based Explanations for Classification over Deterministic and Decomposable Boolean Circuits
https://ojs.aaai.org/index.php/AAAI/article/view/16825
https://ojs.aaai.org/index.php/AAAI/article/download/16825/16632
[ "Marcelo Arenas", "Pablo Barceló", "Leopoldo Bertossi", "Mikaël Monet" ]
Scores based on Shapley values are widely used for providing explanations to classification results over machine learning models. A prime example of this is the influential SHAP-score, a version of the Shapley value that can help explain the result of a learned model on a specific entity by assigning a score to every f...
main
Machine Learning
10.1609/aaai.v35i8.16825
35
8
6670-6678
official
2007.14045
title_judge
10.1609/aaai.v35i8.16826
TabNet: Attentive Interpretable Tabular Learning
https://ojs.aaai.org/index.php/AAAI/article/view/16826
https://ojs.aaai.org/index.php/AAAI/article/download/16826/16633
[ "Sercan Ö. Arik", "Tomas Pfister" ]
We propose a novel high-performance and interpretable canonical deep tabular data learning architecture, TabNet. TabNet uses sequential attention to choose which features to reason from at each decision step, enabling interpretability and more efficient learning as the learning capacity is used for the most salient fea...
main
Machine Learning
10.1609/aaai.v35i8.16826
35
8
6679-6687
official
1908.07442
title_snapshot
10.1609/aaai.v35i8.16827
Robust Model Compression Using Deep Hypotheses
https://ojs.aaai.org/index.php/AAAI/article/view/16827
https://ojs.aaai.org/index.php/AAAI/article/download/16827/16634
[ "Omri Armstrong", "Ran Gilad-Bachrach" ]
Machine Learning models should ideally be compact and robust. Compactness provides efficiency and comprehensibility whereas robustness provides stability. Both topics have been studied in recent years but in isolation. Here we present a robust model compression scheme which is independent of model types: it can compres...
main
Machine Learning
10.1609/aaai.v35i8.16827
35
8
6688-6695
official
2103.07668
title_snapshot
10.1609/aaai.v35i8.16828
Deep Radial-Basis Value Functions for Continuous Control
https://ojs.aaai.org/index.php/AAAI/article/view/16828
https://ojs.aaai.org/index.php/AAAI/article/download/16828/16635
[ "Kavosh Asadi", "Neev Parikh", "Ronald E. Parr", "George D. Konidaris", "Michael L. Littman" ]
A core operation in reinforcement learning (RL) is finding an action that is optimal with respect to a learned value function. This operation is often challenging when the learned value function takes continuous actions as input. We introduce deep radial-basis value functions (RBVFs): value functions learned using a de...
main
Machine Learning
10.1609/aaai.v35i8.16828
35
8
6696-6704
official
2002.01883
title_snapshot
10.1609/aaai.v35i8.16829
DecAug: Out-of-Distribution Generalization via Decomposed Feature Representation and Semantic Augmentation
https://ojs.aaai.org/index.php/AAAI/article/view/16829
https://ojs.aaai.org/index.php/AAAI/article/download/16829/16636
[ "Haoyue Bai", "Rui Sun", "Lanqing Hong", "Fengwei Zhou", "Nanyang Ye", "Han-Jia Ye", "S.-H. Gary Chan", "Zhenguo Li" ]
While deep learning demonstrates its strong ability to handle independent and identically distributed (IID) data, it often suffers from out-of-distribution (OoD) generalization, where the test data come from another distribution (w.r.t. the training one). Designing a general OoD generalization framework for a wide rang...
main
Machine Learning
10.1609/aaai.v35i8.16829
35
8
6705-6713
official
2012.09382
title_snapshot
10.1609/aaai.v35i8.16830
Correlative Channel-Aware Fusion for Multi-View Time Series Classification
https://ojs.aaai.org/index.php/AAAI/article/view/16830
https://ojs.aaai.org/index.php/AAAI/article/download/16830/16637
[ "Yue Bai", "Lichen Wang", "Zhiqiang Tao", "Sheng Li", "Yun Fu" ]
Multi-view time series classification (MVTSC) aims to improve the performance by fusing the distinctive temporal information from multiple views. Existing methods for MVTSC mainly aim to fuse multi-view information at an early stage, e.g., by extracting a common feature subspace among multiple views. However, these app...
main
Machine Learning
10.1609/aaai.v35i8.16830
35
8
6714-6722
official
1911.11561
title_snapshot
10.1609/aaai.v35i8.16831
Deterministic Mini-batch Sequencing for Training Deep Neural Networks
https://ojs.aaai.org/index.php/AAAI/article/view/16831
https://ojs.aaai.org/index.php/AAAI/article/download/16831/16638
[ "Subhankar Banerjee", "Shayok Chakraborty" ]
Recent advancements in the field of deep learning have dramatically improved the performance of machine learning models in a variety of applications, including computer vision, text mining, speech processing and fraud detection among others. Mini-batch gradient descent is the standard algorithm to train deep models, wh...
main
Machine Learning
10.1609/aaai.v35i8.16831
35
8
6723-6731
official
null
null
10.1609/aaai.v35i8.16832
Relative Variational Intrinsic Control
https://ojs.aaai.org/index.php/AAAI/article/view/16832
https://ojs.aaai.org/index.php/AAAI/article/download/16832/16639
[ "Kate Baumli", "David Warde-Farley", "Steven Hansen", "Volodymyr Mnih" ]
In the absence of external rewards, agents can still learn useful behaviors by identifying and mastering a set of diverse skills within their environment. Existing skill learning methods use mutual information objectives to incentivize each skill to be diverse and distinguishable from the rest. However, if care is not ...
main
Machine Learning
10.1609/aaai.v35i8.16832
35
8
6732-6740
official
2012.07827
title_snapshot
10.1609/aaai.v35i8.16833
A Theory of Independent Mechanisms for Extrapolation in Generative Models
https://ojs.aaai.org/index.php/AAAI/article/view/16833
https://ojs.aaai.org/index.php/AAAI/article/download/16833/16640
[ "Michel Besserve", "Remy Sun", "Dominik Janzing", "Bernhard Schölkopf" ]
Generative models can be trained to emulate complex empirical data, but are they useful to make predictions in the context of previously unobserved environments? An intuitive idea to promote such extrapolation capabilities is to have the architecture of such model reflect a causal graph of the true data generating proc...
main
Machine Learning
10.1609/aaai.v35i8.16833
35
8
6741-6749
official
2004.00184
title_snapshot
10.1609/aaai.v35i8.16834
ExGAN: Adversarial Generation of Extreme Samples
https://ojs.aaai.org/index.php/AAAI/article/view/16834
https://ojs.aaai.org/index.php/AAAI/article/download/16834/16641
[ "Siddharth Bhatia", "Arjit Jain", "Bryan Hooi" ]
Mitigating the risk arising from extreme events is a fundamental goal with many applications, such as the modelling of natural disasters, financial crashes, epidemics, and many others. To manage this risk, a vital step is to be able to understand or generate a wide range of extreme scenarios. Existing approaches based ...
main
Machine Learning
10.1609/aaai.v35i8.16834
35
8
6750-6758
official
2009.08454
title_snapshot
10.1609/aaai.v35i8.16835
Ordinal Historical Dependence in Graphical Event Models with Tree Representations
https://ojs.aaai.org/index.php/AAAI/article/view/16835
https://ojs.aaai.org/index.php/AAAI/article/download/16835/16642
[ "Debarun Bhattacharjya", "Tian Gao", "Dharmashankar Subramanian" ]
Graphical event models are representations that capture process independence between different types of events in multivariate temporal point processes. The literature consists of various parametric models and approaches to learn them from multivariate event stream data. Since these models are interpretable, they are o...
main
Machine Learning
10.1609/aaai.v35i8.16835
35
8
6759-6767
official
null
null
10.1609/aaai.v35i8.16836
Characterizing the Loss Landscape in Non-Negative Matrix Factorization
https://ojs.aaai.org/index.php/AAAI/article/view/16836
https://ojs.aaai.org/index.php/AAAI/article/download/16836/16643
[ "Johan Bjorck", "Anmol Kabra", "Kilian Q. Weinberger", "Carla Gomes" ]
Non-negative matrix factorization (NMF) is a highly celebrated algorithm for matrix decomposition that guarantees non-negative factors. The underlying optimization problem is computationally intractable, yet in practice, gradient-descent-based methods often find good solutions. In this paper, we revisit the NMF optimiz...
main
Machine Learning
10.1609/aaai.v35i8.16836
35
8
6768-6776
official
null
null
10.1609/aaai.v35i8.16837
Understanding Decoupled and Early Weight Decay
https://ojs.aaai.org/index.php/AAAI/article/view/16837
https://ojs.aaai.org/index.php/AAAI/article/download/16837/16644
[ "Johan Bjorck", "Kilian Q. Weinberger", "Carla Gomes" ]
Weight decay (WD) is a traditional regularization technique in deep learning, but despite its ubiquity, its behavior is still an area of active research. Golatkar et al. have recently shown that WD only matters at the start of the training in computer vision, upending traditional wisdom. Loshchilov et al. show that for...
main
Machine Learning
10.1609/aaai.v35i8.16837
35
8
6777-6785
official
2012.13841
title_snapshot
10.1609/aaai.v35i8.16838
Communication-Aware Collaborative Learning
https://ojs.aaai.org/index.php/AAAI/article/view/16838
https://ojs.aaai.org/index.php/AAAI/article/download/16838/16645
[ "Avrim Blum", "Shelby Heinecke", "Lev Reyzin" ]
Algorithms for noiseless collaborative PAC learning have been analyzed and optimized in recent years with respect to sample complexity. In this paper, we study collaborative PAC learning with the goal of reducing communication cost at essentially no penalty to the sample complexity. We develop communication efficient c...
main
Machine Learning
10.1609/aaai.v35i8.16838
35
8
6786-6793
official
2012.10569
title_snapshot
10.1609/aaai.v35i8.16839
Stochastic Precision Ensemble: Self-Knowledge Distillation for Quantized Deep Neural Networks
https://ojs.aaai.org/index.php/AAAI/article/view/16839
https://ojs.aaai.org/index.php/AAAI/article/download/16839/16646
[ "Yoonho Boo", "Sungho Shin", "Jungwook Choi", "Wonyong Sung" ]
The quantization of deep neural networks (QDNNs) has been actively studied for deployment in edge devices. Recent studies employ the knowledge distillation (KD) method to improve the performance of quantized networks. In this study, we propose stochastic precision ensemble training for QDNNs (SPEQ). SPEQ is a knowledge...
main
Machine Learning
10.1609/aaai.v35i8.16839
35
8
6794-6802
official
2009.14502
title_snapshot
10.1609/aaai.v35i8.16840
Fast Training of Provably Robust Neural Networks by SingleProp
https://ojs.aaai.org/index.php/AAAI/article/view/16840
https://ojs.aaai.org/index.php/AAAI/article/download/16840/16647
[ "Akhilan Boopathy", "Lily Weng", "Sijia Liu", "Pin-Yu Chen", "Gaoyuan Zhang", "Luca Daniel" ]
Recent works have developed several methods of defending neural networks against adversarial attacks with certified guarantees. However, these techniques can be computationally costly due to the use of certification during training. We develop a new regularizer that is both more efficient than existing certified defens...
main
Machine Learning
10.1609/aaai.v35i8.16840
35
8
6803-6811
official
2102.01208
title_snapshot
10.1609/aaai.v35i8.16841
Sample-Specific Output Constraints for Neural Networks
https://ojs.aaai.org/index.php/AAAI/article/view/16841
https://ojs.aaai.org/index.php/AAAI/article/download/16841/16648
[ "Mathis Brosowsky", "Florian Keck", "Olaf Dünkel", "Marius Zöllner" ]
It is common practice to constrain the output space of a neural network with the final layer to a problem-specific value range. However, for many tasks it is desired to restrict the output space for each input independently to a different subdomain with a non-trivial geometry, e.g. in safety-critical applications, to e...
main
Machine Learning
10.1609/aaai.v35i8.16841
35
8
6812-6821
official
2003.10258
title_snapshot
10.1609/aaai.v35i8.16842
Fairness, Semi-Supervised Learning, and More: A General Framework for Clustering with Stochastic Pairwise Constraints
https://ojs.aaai.org/index.php/AAAI/article/view/16842
https://ojs.aaai.org/index.php/AAAI/article/download/16842/16649
[ "Brian Brubach", "Darshan Chakrabarti", "John P. Dickerson", "Aravind Srinivasan", "Leonidas Tsepenekas" ]
Metric clustering is fundamental in areas ranging from Combinatorial Optimization and Data Mining, to Machine Learning and Operations Research. However, in a variety of situations we may have additional requirements or knowledge, distinct from the underlying metric, regarding which pairs of points should be clustered t...
main
Machine Learning
10.1609/aaai.v35i8.16842
35
8
6822-6830
official
2103.02013
title_snapshot
10.1609/aaai.v35i8.16811
SWIFT: Scalable Wasserstein Factorization for Sparse Nonnegative Tensors
https://ojs.aaai.org/index.php/AAAI/article/view/16811
https://ojs.aaai.org/index.php/AAAI/article/download/16811/16618
[ "Ardavan Afshar", "Kejing Yin", "Sherry Yan", "Cheng Qian", "Joyce Ho", "Haesun Park", "Jimeng Sun" ]
Existing tensor factorization methods assume that the input tensor follows some specific distribution (i.e. Poisson, Bernoulli, and Gaussian), and solve the factorization by minimizing some empirical loss functions defined based on the corresponding distribution. However, it suffers from several drawbacks: 1) In realit...
main
Machine Learning
10.1609/aaai.v35i8.16811
35
8
6548-6556
official
2010.04081
title_snapshot
10.1609/aaai.v35i8.16812
DART: Adaptive Accept Reject Algorithm for Non-Linear Combinatorial Bandits
https://ojs.aaai.org/index.php/AAAI/article/view/16812
https://ojs.aaai.org/index.php/AAAI/article/download/16812/16619
[ "Mridul Agarwal", "Vaneet Aggarwal", "Abhishek Kumar Umrawal", "Chris Quinn" ]
We consider the bandit problem of selecting K out of N arms at each time step. The joint reward can be a non-linear function of the rewards of the selected individual arms. The direct use of a multi-armed bandit algorithm requires choosing among all possible combinations, making the action space large. To simplify the ...
main
Machine Learning
10.1609/aaai.v35i8.16812
35
8
6557-6565
official
null
null
10.1609/aaai.v35i8.16813
Improved Worst-Case Regret Bounds for Randomized Least-Squares Value Iteration
https://ojs.aaai.org/index.php/AAAI/article/view/16813
https://ojs.aaai.org/index.php/AAAI/article/download/16813/16620
[ "Priyank Agrawal", "Jinglin Chen", "Nan Jiang" ]
This paper studies regret minimization with randomized value functions in reinforcement learning. In tabular finite-horizon Markov Decision Processes, we introduce a clipping variant of one classical Thompson Sampling (TS)-like algorithm, randomized least-squares value iteration (RLSVI). Our $\tilde{\mathrm{O}}(H^2S\sq...
main
Machine Learning
10.1609/aaai.v35i8.16813
35
8
6566-6573
official
2010.12163
title_snapshot
10.1609/aaai.v35i8.16814
Semi-supervised Sequence Classification through Change Point Detection
https://ojs.aaai.org/index.php/AAAI/article/view/16814
https://ojs.aaai.org/index.php/AAAI/article/download/16814/16621
[ "Nauman Ahad", "Mark A. Davenport" ]
Sequential sensor data is generated in a wide variety of real-world applications. A fundamental machine learning challenge involves learning effective classifiers for such sequential data. While deep learning has led to impressive performance gains in recent years within domains such as speech, this has relied on the a...
main
Machine Learning
10.1609/aaai.v35i8.16814
35
8
6574-6581
official
2009.11829
title_snapshot
10.1609/aaai.v35i8.16815
Learning Invariant Representations using Inverse Contrastive Loss
https://ojs.aaai.org/index.php/AAAI/article/view/16815
https://ojs.aaai.org/index.php/AAAI/article/download/16815/16622
[ "Aditya Kumar Akash", "Vishnu Suresh Lokhande", "Sathya N. Ravi", "Vikas Singh" ]
Learning invariant representations is a critical first step in a number of machine learning tasks. A common approach is given by the so-called information bottleneck principle in which an application dependent function of mutual information is carefully chosen and optimized. Unfortunately, in practice, these functions ...
main
Machine Learning
10.1609/aaai.v35i8.16815
35
8
6582-6591
official
2102.08343
title_snapshot
10.1609/aaai.v35i8.16816
Learned Bi-Resolution Image Coding using Generalized Octave Convolutions
https://ojs.aaai.org/index.php/AAAI/article/view/16816
https://ojs.aaai.org/index.php/AAAI/article/download/16816/16623
[ "Mohammad Akbari", "Jie Liang", "Jingning Han", "Chengjie Tu" ]
Learned image compression has recently shown the potential to outperform the standard codecs. State-of-the-art rate-distortion (R-D) performance has been achieved by context-adaptive entropy coding approaches in which hyperprior and autoregressive models are jointly utilized to effectively capture the spatial dependenc...
main
Machine Learning
10.1609/aaai.v35i8.16816
35
8
6592-6599
official
null
null
10.1609/aaai.v35i8.16817
Deep Bayesian Quadrature Policy Optimization
https://ojs.aaai.org/index.php/AAAI/article/view/16817
https://ojs.aaai.org/index.php/AAAI/article/download/16817/16624
[ "Ravi Tej Akella", "Kamyar Azizzadenesheli", "Mohammad Ghavamzadeh", "Animashree Anandkumar", "Yisong Yue" ]
We study the problem of obtaining accurate policy gradient estimates using a finite number of samples. Monte-Carlo methods have been the default choice for policy gradient estimation, despite suffering from high variance in the gradient estimates. On the other hand, more sample efficient alternatives like Bayesian quad...
main
Machine Learning
10.1609/aaai.v35i8.16817
35
8
6600-6608
official
2006.15637
title_snapshot
10.1609/aaai.v35i8.16818
eTREE: Learning Tree-structured Embeddings
https://ojs.aaai.org/index.php/AAAI/article/view/16818
https://ojs.aaai.org/index.php/AAAI/article/download/16818/16625
[ "Faisal M. Almutairi", "Yunlong Wang", "Dong Wang", "Emily Zhao", "Nicholas D. Sidiropoulos" ]
Matrix factorization (MF) plays an important role in a wide range of machine learning and data mining models. MF is commonly used to obtain item embeddings and feature representations due to its ability to capture correlations and higher-order statistical dependencies across dimensions. In many applications, the catego...
main
Machine Learning
10.1609/aaai.v35i8.16818
35
8
6609-6617
official
2012.10853
title_snapshot
10.1609/aaai.v35i8.16819
Does Explainable Artificial Intelligence Improve Human Decision-Making?
https://ojs.aaai.org/index.php/AAAI/article/view/16819
https://ojs.aaai.org/index.php/AAAI/article/download/16819/16626
[ "Yasmeen Alufaisan", "Laura R. Marusich", "Jonathan Z. Bakdash", "Yan Zhou", "Murat Kantarcioglu" ]
Explainable AI provides insights to users into the why for model predictions, offering potential for users to better understand and trust a model, and to recognize and correct AI predictions that are incorrect. Prior research on human and explainable AI interactions has focused on measures such as interpretability, tru...
main
Machine Learning
10.1609/aaai.v35i8.16819
35
8
6618-6626
official
2006.11194
title_snapshot
10.1609/aaai.v35i8.16820
Decentralized Multi-Agent Linear Bandits with Safety Constraints
https://ojs.aaai.org/index.php/AAAI/article/view/16820
https://ojs.aaai.org/index.php/AAAI/article/download/16820/16627
[ "Sanae Amani", "Christos Thrampoulidis" ]
We study decentralized stochastic linear bandits, where a network of N agents acts cooperatively to efficiently solve a linear bandit-optimization problem over a d-dimensional space. For this problem, we propose DLUCB: a fully decentralized algorithm that minimizes the cumulative regret over the entire network. At each...
main
Machine Learning
10.1609/aaai.v35i8.16820
35
8
6627-6635
official
2012.00314
title_snapshot
10.1609/aaai.v35i8.16821
Computing an Efficient Exploration Basis for Learning with Univariate Polynomial Features
https://ojs.aaai.org/index.php/AAAI/article/view/16821
https://ojs.aaai.org/index.php/AAAI/article/download/16821/16628
[ "Chaitanya Amballa", "Manu K. Gupta", "Sanjay P. Bhat" ]
Barycentric spanners have been used as an efficient exploration basis in online linear optimization problems in a bandit framework. We characterise the barycentric spanner for decision problems in which the cost (or reward) is a polynomial in a single decision variable. Our characterisation of the barycentric spanner i...
main
Machine Learning
10.1609/aaai.v35i8.16821
35
8
6636-6643
official
null
null
10.1609/aaai.v35i8.16822
Noise Estimation Using Density Estimation for Self-Supervised Multimodal Learning
https://ojs.aaai.org/index.php/AAAI/article/view/16822
https://ojs.aaai.org/index.php/AAAI/article/download/16822/16629
[ "Elad Amrani", "Rami Ben-Ari", "Daniel Rotman", "Alex Bronstein" ]
One of the key factors of enabling machine learning models to comprehend and solve real-world tasks is to leverage multimodal data. Unfortunately, annotation of multimodal data is challenging and expensive. Recently, self-supervised multimodal methods that combine vision and language were proposed to learn multimodal r...
main
Machine Learning
10.1609/aaai.v35i8.16822
35
8
6644-6652
official
2003.03186
title_snapshot
10.1609/aaai.v35i9.17003
Hypothesis Disparity Regularized Mutual Information Maximization
https://ojs.aaai.org/index.php/AAAI/article/view/17003
https://ojs.aaai.org/index.php/AAAI/article/download/17003/16810
[ "Qicheng Lao", "Xiang Jiang", "Mohammad Havaei" ]
We propose a hypothesis disparity regularized mutual information maximization (HDMI) approach to tackle unsupervised hypothesis transfer---as an effort towards unifying hypothesis transfer learning (HTL) and unsupervised domain adaptation (UDA)---where the knowledge from a source domain is transferred solely through hy...
main
Machine Learning
10.1609/aaai.v35i9.17003
35
9
8243-8251
official
2012.08072
title_snapshot
10.1609/aaai.v35i9.17004
Query Training: Learning a Worse Model to Infer Better Marginals in Undirected Graphical Models with Hidden Variables
https://ojs.aaai.org/index.php/AAAI/article/view/17004
https://ojs.aaai.org/index.php/AAAI/article/download/17004/16811
[ "Miguel Lázaro-Gredilla", "Wolfgang Lehrach", "Nishad Gothoskar", "Guangyao Zhou", "Antoine Dedieu", "Dileep George" ]
Probabilistic graphical models (PGMs) provide a compact representation of knowledge that can be queried in a flexible way: after learning the parameters of a graphical model once, new probabilistic queries can be answered at test time without retraining. However, when using undirected PGMS with hidden variables, two so...
main
Machine Learning
10.1609/aaai.v35i9.17004
35
9
8252-8260
official
2006.06803
title_snapshot
10.1609/aaai.v35i9.17005
Metrics and Continuity in Reinforcement Learning
https://ojs.aaai.org/index.php/AAAI/article/view/17005
https://ojs.aaai.org/index.php/AAAI/article/download/17005/16812
[ "Charline Le Lan", "Marc G. Bellemare", "Pablo Samuel Castro" ]
In most practical applications of reinforcement learning, it is untenable to maintain direct estimates for individual states; in continuous-state systems, it is impossible. Instead, researchers often leverage {\em state similarity} (whether explicitly or implicitly) to build models that can generalize well from a limit...
main
Machine Learning
10.1609/aaai.v35i9.17005
35
9
8261-8269
official
2102.01514
title_snapshot
10.1609/aaai.v35i9.17006
Lipschitz Lifelong Reinforcement Learning
https://ojs.aaai.org/index.php/AAAI/article/view/17006
https://ojs.aaai.org/index.php/AAAI/article/download/17006/16813
[ "Erwan Lecarpentier", "David Abel", "Kavosh Asadi", "Yuu Jinnai", "Emmanuel Rachelson", "Michael L. Littman" ]
We consider the problem of knowledge transfer when an agent is facing a series of Reinforcement Learning (RL) tasks. We introduce a novel metric between Markov Decision Processes and establish that close MDPs have close optimal value functions. Formally, the optimal value functions are Lipschitz continuous with respect...
main
Machine Learning
10.1609/aaai.v35i9.17006
35
9
8270-8278
official
2001.05411
title_snapshot
10.1609/aaai.v35i9.17007
Norm-Based Generalisation Bounds for Deep Multi-Class Convolutional Neural Networks
https://ojs.aaai.org/index.php/AAAI/article/view/17007
https://ojs.aaai.org/index.php/AAAI/article/download/17007/16814
[ "Antoine Ledent", "Waleed Mustafa", "Yunwen Lei", "Marius Kloft" ]
We show generalisation error bounds for deep learning with two main improvements over the state of the art. (1) Our bounds have no explicit dependence on the number of classes except for logarithmic factors. This holds even when formulating the bounds in terms of the Frobenius-norm of the weight matrices, where previou...
main
Machine Learning
10.1609/aaai.v35i9.17007
35
9
8279-8287
official
1905.12430
title_judge
10.1609/aaai.v35i9.17008
Learnable Dynamic Temporal Pooling for Time Series Classification
https://ojs.aaai.org/index.php/AAAI/article/view/17008
https://ojs.aaai.org/index.php/AAAI/article/download/17008/16815
[ "Dongha Lee", "Seonghyeon Lee", "Hwanjo Yu" ]
With the increase of available time series data, predicting their class labels has been one of the most important challenges in a wide range of disciplines. Recent studies on time series classification show that convolutional neural networks (CNN) achieved the state-of-the-art performance as a single classifier. In thi...
main
Machine Learning
10.1609/aaai.v35i9.17008
35
9
8288-8296
official
2104.02577
title_snapshot
10.1609/aaai.v35i9.17009
Interpretable Embedding Procedure Knowledge Transfer via Stacked Principal Component Analysis and Graph Neural Network
https://ojs.aaai.org/index.php/AAAI/article/view/17009
https://ojs.aaai.org/index.php/AAAI/article/download/17009/16816
[ "Seunghyun Lee", "Byung Cheol Song" ]
Knowledge distillation (KD) is one of the most useful techniques for light-weight neural networks. Although neural networks have a clear purpose of embedding datasets into the low-dimensional space, the existing knowledge was quite far from this purpose and provided only limited information. We argue that good knowledg...
main
Machine Learning
10.1609/aaai.v35i9.17009
35
9
8297-8305
official
2104.13561
title_snapshot
10.1609/aaai.v35i9.17010
Unsupervised Domain Adaptation for Semantic Segmentation by Content Transfer
https://ojs.aaai.org/index.php/AAAI/article/view/17010
https://ojs.aaai.org/index.php/AAAI/article/download/17010/16817
[ "Suhyeon Lee", "Junhyuk Hyun", "Hongje Seong", "Euntai Kim" ]
In this paper, we tackle the unsupervised domain adaptation (UDA) for semantic segmentation, which aims to segment the unlabeled real data using labeled synthetic data. The main problem of UDA for semantic segmentation relies on reducing the domain gap between the real image and synthetic image. To solve this problem, ...
main
Machine Learning
10.1609/aaai.v35i9.17010
35
9
8306-8315
official
2012.12545
title_snapshot
10.1609/aaai.v35i9.17011
Memory and Computation-Efficient Kernel SVM via Binary Embedding and Ternary Model Coefficients
https://ojs.aaai.org/index.php/AAAI/article/view/17011
https://ojs.aaai.org/index.php/AAAI/article/download/17011/16818
[ "Zijian Lei", "Liang Lan" ]
Kernel approximation is widely used to scale up kernel SVM training and prediction. However, the memory and computation costs of kernel approximation models are still too large if we want to deploy them on memory-limited devices such as mobile phones, smart watches and IoT devices. To address this challenge, we propose...
main
Machine Learning
10.1609/aaai.v35i9.17011
35
9
8316-8323
official
2010.02577
title_snapshot
10.1609/aaai.v35i9.17012
Enhancing Parameter-Free Frank Wolfe with an Extra Subproblem
https://ojs.aaai.org/index.php/AAAI/article/view/17012
https://ojs.aaai.org/index.php/AAAI/article/download/17012/16819
[ "Bingcong Li", "Lingda Wang", "Georgios B. Giannakis", "Zhizhen Zhao" ]
Aiming at convex optimization under structural constraints, this work introduces and analyzes a variant of the Frank Wolfe (FW) algorithm termed ExtraFW. The distinct feature of ExtraFW is the pair of gradients leveraged per iteration, thanks to which the decision variable is updated in a prediction-correction (PC) for...
main
Machine Learning
10.1609/aaai.v35i9.17012
35
9
8324-8331
official
2012.05284
title_snapshot
10.1609/aaai.v35i9.17013
Unsupervised Active Learning via Subspace Learning
https://ojs.aaai.org/index.php/AAAI/article/view/17013
https://ojs.aaai.org/index.php/AAAI/article/download/17013/16820
[ "Changsheng Li", "Kaihang Mao", "Lingyan Liang", "Dongchun Ren", "Wei Zhang", "Ye Yuan", "Guoren Wang" ]
Unsupervised active learning has been an active research topic in machine learning community, with the purpose of choosing representative samples to be labelled in an unsupervised manner. Previous works usually take the minimization of data reconstruction loss as the criterion to select representative samples which can...
main
Machine Learning
10.1609/aaai.v35i9.17013
35
9
8332-8339
official
null
null
10.1609/aaai.v35i9.17014
LRSC: Learning Representations for Subspace Clustering
https://ojs.aaai.org/index.php/AAAI/article/view/17014
https://ojs.aaai.org/index.php/AAAI/article/download/17014/16821
[ "Changsheng Li", "Chen Yang", "Bo Liu", "Ye Yuan", "Guoren Wang" ]
Deep learning based subspace clustering methods have attracted increasing attention in recent years, where a basic theme is to non-linearly map data into a latent space, and then uncover subspace structures based upon the data self-expressiveness property. However, almost all existing deep subspace clustering methods o...
main
Machine Learning
10.1609/aaai.v35i9.17014
35
9
8340-8348
official
null
null
10.1609/aaai.v35i9.17015
GoT: a Growing Tree Model for Clustering Ensemble
https://ojs.aaai.org/index.php/AAAI/article/view/17015
https://ojs.aaai.org/index.php/AAAI/article/download/17015/16822
[ "Feijiang Li", "Yuhua Qian", "Jieting Wang" ]
The clustering ensemble technique that integrates multiple clustering results can improve the accuracy and robustness of the final clustering. In many clustering ensemble algorithms, the co-association matrix (CA matrix), which reflects the frequency of any two samples being partitioned into the same cluster, plays an ...
main
Machine Learning
10.1609/aaai.v35i9.17015
35
9
8349-8356
official
null
null
10.1609/aaai.v35i9.17016
VSQL: Variational Shadow Quantum Learning for Classification
https://ojs.aaai.org/index.php/AAAI/article/view/17016
https://ojs.aaai.org/index.php/AAAI/article/download/17016/16823
[ "Guangxi Li", "Zhixin Song", "Xin Wang" ]
Classification of quantum data is essential for quantum machine learning and near-term quantum technologies. In this paper, we propose a new hybrid quantum-classical framework for supervised quantum learning, which we call Variational Shadow Quantum Learning (VSQL). Our method in particular utilizes the classical shado...
main
Machine Learning
10.1609/aaai.v35i9.17016
35
9
8357-8365
official
2012.08288
title_snapshot
10.1609/aaai.v35i9.17017
High Fidelity GAN Inversion via Prior Multi-Subspace Feature Composition
https://ojs.aaai.org/index.php/AAAI/article/view/17017
https://ojs.aaai.org/index.php/AAAI/article/download/17017/16824
[ "Guanyue Li", "Qianfen Jiao", "Sheng Qian", "Si Wu", "Hau-San Wong" ]
Generative Adversarial Networks (GANs) have shown impressive gains in image synthesis. GAN inversion was recently studied to understand and utilize the knowledge it learns, where a real image is inverted back to a latent code and can thus be reconstructed by the generator. Although increasing the number of latent codes...
main
Machine Learning
10.1609/aaai.v35i9.17017
35
9
8366-8374
official
null
null
10.1609/aaai.v35i9.17018
ShapeNet: A Shapelet-Neural Network Approach for Multivariate Time Series Classification
https://ojs.aaai.org/index.php/AAAI/article/view/17018
https://ojs.aaai.org/index.php/AAAI/article/download/17018/16825
[ "Guozhong Li", "Byron Choi", "Jianliang Xu", "Sourav S Bhowmick", "Kwok-Pan Chun", "Grace Lai-Hung Wong" ]
Time series shapelets are short discriminative subsequences that recently have been found not only to be accurate but also interpretable for the classification problem of univariate time series (UTS). However, existing work on shapelets selection cannot be applied to multivariate time series classification (MTSC) since...
main
Machine Learning
10.1609/aaai.v35i9.17018
35
9
8375-8383
official
null
null
10.1609/aaai.v35i9.17019
A Bayesian Approach for Subset Selection in Contextual Bandits
https://ojs.aaai.org/index.php/AAAI/article/view/17019
https://ojs.aaai.org/index.php/AAAI/article/download/17019/16826
[ "Jialian Li", "Chao Du", "Jun Zhu" ]
Subset selection in Contextual Bandits (CB) is an important task in various applications such as advertisement recommendation. In CB, arms are attached with contexts and thus correlated in the context space. Proper exploration for subset selection in CB should carefully consider the contexts. Previous works mainly conc...
main
Machine Learning
10.1609/aaai.v35i9.17019
35
9
8384-8391
official
null
null
10.1609/aaai.v35i9.17020
Self-Paced Two-dimensional PCA
https://ojs.aaai.org/index.php/AAAI/article/view/17020
https://ojs.aaai.org/index.php/AAAI/article/download/17020/16827
[ "Jiangxin Li", "Zhao Kang", "Chong Peng", "Wenyu Chen" ]
Two-dimensional PCA (2DPCA) is an effective approach to reduce dimension and extract features in the image domain. Most recently developed techniques use different error measures to improve their robustness to outliers. When certain data points are overly contaminated, the existing methods are frequently incapable of f...
main
Machine Learning
10.1609/aaai.v35i9.17020
35
9
8392-8400
official
null
null
10.1609/aaai.v35i9.17021
Learning Intact Features by Erasing-Inpainting for Few-shot Classification
https://ojs.aaai.org/index.php/AAAI/article/view/17021
https://ojs.aaai.org/index.php/AAAI/article/download/17021/16828
[ "Junjie Li", "Zilei Wang", "Xiaoming Hu" ]
Few-shot classification aims to categorize the samples from unseen classes with only few labeled samples. To address such a challenge, many methods exploit a base set consisting of massive labeled samples to learn an instance embedding function, i.e., image feature extractor, and it is expected to possess good transfer...
main
Machine Learning
10.1609/aaai.v35i9.17021
35
9
8401-8409
official
null
null
10.1609/aaai.v35i9.17022
Token-Aware Virtual Adversarial Training in Natural Language Understanding
https://ojs.aaai.org/index.php/AAAI/article/view/17022
https://ojs.aaai.org/index.php/AAAI/article/download/17022/16829
[ "Linyang Li", "Xipeng Qiu" ]
Gradient-based adversarial training is widely used in improving the robustness of neural networks, while it cannot be easily adapted to natural language processing tasks since the embedding space is discrete. In natural language processing fields, virtual adversarial training is introduced since texts are discrete and ...
main
Machine Learning
10.1609/aaai.v35i9.17022
35
9
8410-8418
official
2004.14543
title_judge
10.1609/aaai.v35i9.16983
Learning Generalized Relational Heuristic Networks for Model-Agnostic Planning
https://ojs.aaai.org/index.php/AAAI/article/view/16983
https://ojs.aaai.org/index.php/AAAI/article/download/16983/16790
[ "Rushang Karia", "Siddharth Srivastava" ]
Computing goal-directed behavior is essential to designing efficient AI systems. Due to the computational complexity of planning, current approaches rely primarily upon hand-coded symbolic action models and hand-coded heuristic function generators for efficiency. Learned heuristics for such problems have been of limite...
main
Machine Learning
10.1609/aaai.v35i9.16983
35
9
8064-8073
official
2007.06702
title_snapshot
10.1609/aaai.v35i9.16984
A Recipe for Global Convergence Guarantee in Deep Neural Networks
https://ojs.aaai.org/index.php/AAAI/article/view/16984
https://ojs.aaai.org/index.php/AAAI/article/download/16984/16791
[ "Kenji Kawaguchi", "Qingyun Sun" ]
Existing global convergence guarantees of (stochastic) gradient descent do not apply to practical deep networks in the practical regime of deep learning beyond the neural tangent kernel (NTK) regime. This paper proposes an algorithm, which is ensured to have global convergence guarantees in the practical regime beyond ...
main
Machine Learning
10.1609/aaai.v35i9.16984
35
9
8074-8082
official
2104.05785
title_snapshot
10.1609/aaai.v35i9.16985
Bayesian Dynamic Mode Decomposition with Variational Matrix Factorization
https://ojs.aaai.org/index.php/AAAI/article/view/16985
https://ojs.aaai.org/index.php/AAAI/article/download/16985/16792
[ "Takahiro Kawashima", "Hayaru Shouno", "Hideitsu Hino" ]
Dynamic mode decomposition (DMD) and its extensions are data-driven methods that have substantially contributed to our understanding of dynamical systems. However, because DMD and most of its extensions are deterministic, it is difficult to treat probabilistic representations of parameters and predictions. In this work...
main
Machine Learning
10.1609/aaai.v35i9.16985
35
9
8083-8091
official
null
null
10.1609/aaai.v35i9.16986
Improving Fairness and Privacy in Selection Problems
https://ojs.aaai.org/index.php/AAAI/article/view/16986
https://ojs.aaai.org/index.php/AAAI/article/download/16986/16793
[ "Mohammad Mahdi Khalili", "Xueru Zhang", "Mahed Abroshan", "Somayeh Sojoudi" ]
Supervised learning models have been increasingly used for making decisions about individuals in applications such as hiring, lending, and college admission. These models may inherit pre-existing biases from training datasets and discriminate against protected attributes (e.g., race or gender). In addition to unfairnes...
main
Machine Learning
10.1609/aaai.v35i9.16986
35
9
8092-8100
official
2012.03812
title_snapshot
10.1609/aaai.v35i9.16987
A Flexible Framework for Communication-Efficient Machine Learning
https://ojs.aaai.org/index.php/AAAI/article/view/16987
https://ojs.aaai.org/index.php/AAAI/article/download/16987/16794
[ "Sarit Khirirat", "Sindri Magnússon", "Arda Aytekin", "Mikael Johansson" ]
With the increasing scale of machine learning tasks, it has become essential to reduce the communication between computing nodes. Early work on gradient compression focused on the bottleneck between CPUs and GPUs, but communication-efficiency is now needed in a variety of different system architectures, from high-perfo...
main
Machine Learning
10.1609/aaai.v35i9.16987
35
9
8101-8109
official
null
null
10.1609/aaai.v35i9.16988
GLISTER: Generalization based Data Subset Selection for Efficient and Robust Learning
https://ojs.aaai.org/index.php/AAAI/article/view/16988
https://ojs.aaai.org/index.php/AAAI/article/download/16988/16795
[ "Krishnateja Killamsetty", "Durga Sivasubramanian", "Ganesh Ramakrishnan", "Rishabh Iyer" ]
Large scale machine learning and deep models are extremely data-hungry. Unfortunately, obtaining large amounts of labeled data is expensive, and training state-of-the-art models (with hyperparameter tuning) requires significant computing resources and time. Secondly, real-world data is noisy and imbalanced. As a result...
main
Machine Learning
10.1609/aaai.v35i9.16988
35
9
8110-8118
official
2012.10630
title_snapshot
10.1609/aaai.v35i9.16989
Understanding Catastrophic Overfitting in Single-step Adversarial Training
https://ojs.aaai.org/index.php/AAAI/article/view/16989
https://ojs.aaai.org/index.php/AAAI/article/download/16989/16796
[ "Hoki Kim", "Woojin Lee", "Jaewook Lee" ]
Although fast adversarial training has demonstrated both robustness and efficiency, the problem of "catastrophic overfitting" has been observed. This is a phenomenon in which, during single-step adversarial training, the robust accuracy against projected gradient descent (PGD) suddenly decreases to 0% after a few epoch...
main
Machine Learning
10.1609/aaai.v35i9.16989
35
9
8119-8127
official
2010.01799
title_snapshot
10.1609/aaai.v35i9.16990
Counterfactual Fairness with Disentangled Causal Effect Variational Autoencoder
https://ojs.aaai.org/index.php/AAAI/article/view/16990
https://ojs.aaai.org/index.php/AAAI/article/download/16990/16797
[ "Hyemi Kim", "Seungjae Shin", "JoonHo Jang", "Kyungwoo Song", "Weonyoung Joo", "Wanmo Kang", "Il-Chul Moon" ]
The problem of fair classification can be mollified if we develop a method to remove the embedded sensitive information from the classification features. This line of separating the sensitive information is developed through the causal inference, and the causal inference enables the counterfactual generations to contra...
main
Machine Learning
10.1609/aaai.v35i9.16990
35
9
8128-8136
official
2011.11878
title_snapshot
10.1609/aaai.v35i9.16991
Split-and-Bridge: Adaptable Class Incremental Learning within a Single Neural Network
https://ojs.aaai.org/index.php/AAAI/article/view/16991
https://ojs.aaai.org/index.php/AAAI/article/download/16991/16798
[ "Jong-Yeong Kim", "Dong-Wan Choi" ]
Continual learning has been a major problem in the deep learning community, where the main challenge is how to effectively learn a series of newly arriving tasks without forgetting the knowledge of previous tasks. Initiated by Learning without Forgetting (LwF), many of the existing works report that knowledge distillat...
main
Machine Learning
10.1609/aaai.v35i9.16991
35
9
8137-8145
official
2107.01349
title_snapshot
10.1609/aaai.v35i9.16992
DPM: A Novel Training Method for Physics-Informed Neural Networks in Extrapolation
https://ojs.aaai.org/index.php/AAAI/article/view/16992
https://ojs.aaai.org/index.php/AAAI/article/download/16992/16799
[ "Jungeun Kim", "Kookjin Lee", "Dongeun Lee", "Sheo Yon Jhin", "Noseong Park" ]
We present a method for learning dynamics of complex physical processes described by time-dependent nonlinear partial differential equations (PDEs). Our particular interest lies in extrapolating solutions in time beyond the range of temporal domain used in training. Our choice for a baseline method is physics-informed ...
main
Machine Learning
10.1609/aaai.v35i9.16992
35
9
8146-8154
official
2012.02681
title_snapshot
10.1609/aaai.v35i9.16993
Kernel-convoluted Deep Neural Networks with Data Augmentation
https://ojs.aaai.org/index.php/AAAI/article/view/16993
https://ojs.aaai.org/index.php/AAAI/article/download/16993/16800
[ "Minjin Kim", "Young-geun Kim", "Dongha Kim", "Yongdai Kim", "Myunghee Cho Paik" ]
The Mixup method, which uses linearly interpolated data, has emerged as an effective data augmentation tool to improve generalization performance and the robustness to adversarial examples. The motivation is to curtail undesirable oscillations by its implicit model constraint to behave linearly at in-between observed d...
main
Machine Learning
10.1609/aaai.v35i9.16993
35
9
8155-8162
official
2012.02521
title_snapshot
10.1609/aaai.v35i9.16994
Neural Sequence-to-grid Module for Learning Symbolic Rules
https://ojs.aaai.org/index.php/AAAI/article/view/16994
https://ojs.aaai.org/index.php/AAAI/article/download/16994/16801
[ "Segwang Kim", "Hyoungwook Nam", "Joonyoung Kim", "Kyomin Jung" ]
Logical reasoning tasks over symbols, such as learning arithmetic operations and computer program evaluations, have become challenges to deep learning. In particular, even state-of-the-art neural networks fail to achieve \textit{out-of-distribution} (OOD) generalization of symbolic reasoning tasks, whereas humans can e...
main
Machine Learning
10.1609/aaai.v35i9.16994
35
9
8163-8171
official
2101.04921
title_snapshot
10.1609/aaai.v35i9.16995
Visual Concept Reasoning Networks
https://ojs.aaai.org/index.php/AAAI/article/view/16995
https://ojs.aaai.org/index.php/AAAI/article/download/16995/16802
[ "Taesup Kim", "Sungwoong Kim", "Yoshua Bengio" ]
A split-transform-merge strategy has been broadly used as an architectural constraint in convolutional neural networks for visual recognition tasks. It approximates sparsely connected networks by explicitly defining multiple branches to simultaneously learn representations with different visual concepts or properties. ...
main
Machine Learning
10.1609/aaai.v35i9.16995
35
9
8172-8180
official
2008.11783
title_snapshot
10.1609/aaai.v35i9.16996
Sparsity Aware Normalization for GANs
https://ojs.aaai.org/index.php/AAAI/article/view/16996
https://ojs.aaai.org/index.php/AAAI/article/download/16996/16803
[ "Idan Kligvasser", "Tomer Michaeli" ]
Generative adversarial networks (GANs) are known to benefit from regularization or normalization of their critic (discriminator) network during training. In this paper, we analyze the popular spectral normalization scheme, find a significant drawback and introduce sparsity aware normalization (SAN), a new alternative a...
main
Machine Learning
10.1609/aaai.v35i9.16996
35
9
8181-8190
official
2103.02458
title_snapshot
10.1609/aaai.v35i9.16997
HINT: Hierarchical Invertible Neural Transport for Density Estimation and Bayesian Inference
https://ojs.aaai.org/index.php/AAAI/article/view/16997
https://ojs.aaai.org/index.php/AAAI/article/download/16997/16804
[ "Jakob Kruse", "Gianluca Detommaso", "Ullrich Köthe", "Robert Scheichl" ]
Many recent invertible neural architectures are based on coupling block designs where variables are divided in two subsets which serve as inputs of an easily invertible (usually affine) triangular transformation. While such a transformation is invertible, its Jacobian is very sparse and thus may lack expressiveness. Th...
main
Machine Learning
10.1609/aaai.v35i9.16997
35
9
8191-8199
official
1905.10687
title_snapshot
10.1609/aaai.v35i9.16998
Nearly Linear-Time, Parallelizable Algorithms for Non-Monotone Submodular Maximization
https://ojs.aaai.org/index.php/AAAI/article/view/16998
https://ojs.aaai.org/index.php/AAAI/article/download/16998/16805
[ "Alan Kuhnle" ]
We study combinatorial, parallelizable algorithms for maximization of a submodular function, not necessarily monotone, with respect to a cardinality constraint k. We improve the best approximation factor achieved by an algorithm that has optimal adaptivity and query complexity, up to logarithmic factors in the size of ...
main
Machine Learning
10.1609/aaai.v35i9.16998
35
9
8200-8208
official
2009.01947
title_judge
10.1609/aaai.v35i9.16999
Asynchronous Optimization Methods for Efficient Training of Deep Neural Networks with Guarantees
https://ojs.aaai.org/index.php/AAAI/article/view/16999
https://ojs.aaai.org/index.php/AAAI/article/download/16999/16806
[ "Vyacheslav Kungurtsev", "Malcolm Egan", "Bapi Chatterjee", "Dan Alistarh" ]
Asynchronous distributed algorithms are a popular way to reduce synchronization costs in large-scale optimization, and in particular for neural network training. However, for nonsmooth and nonconvex objectives, few convergence guarantees exist beyond cases where closed-form proximal operator solutions are available. As...
main
Machine Learning
10.1609/aaai.v35i9.16999
35
9
8209-8216
official
1905.11845
title_snapshot
10.1609/aaai.v35i9.17000
Positions, Channels, and Layers: Fully Generalized Non-Local Network for Singer Identification
https://ojs.aaai.org/index.php/AAAI/article/view/17000
https://ojs.aaai.org/index.php/AAAI/article/download/17000/16807
[ "I-Yuan Kuo", "Wen-Li Wei", "Jen-Chun Lin" ]
Recently, a non-local (NL) operation has been designed as the central building block for deep-net models to capture long-range dependencies (Wang et al. 2018). Despite its excellent performance, it does not consider the interaction between positions across channels and layers, which is crucial in fine-grained classific...
main
Machine Learning
10.1609/aaai.v35i9.17000
35
9
8217-8225
official
null
null
10.1609/aaai.v35i9.17001
MolGrow: A Graph Normalizing Flow for Hierarchical Molecular Generation
https://ojs.aaai.org/index.php/AAAI/article/view/17001
https://ojs.aaai.org/index.php/AAAI/article/download/17001/16808
[ "Maksim Kuznetsov", "Daniil Polykovskiy" ]
We propose a hierarchical normalizing flow model for generating molecular graphs. The model produces new molecular structures from a single-node graph by recursively splitting every node into two. All operations are invertible and can be used as plug-and-play modules. The hierarchical nature of the latent codes allows ...
main
Machine Learning
10.1609/aaai.v35i9.17001
35
9
8226-8234
official
2106.05856
title_snapshot
10.1609/aaai.v35i9.17002
Compressing Deep Convolutional Neural Networks by Stacking Low-dimensional Binary Convolution Filters
https://ojs.aaai.org/index.php/AAAI/article/view/17002
https://ojs.aaai.org/index.php/AAAI/article/download/17002/16809
[ "Weichao Lan", "Liang Lan" ]
Deep Convolutional Neural Networks (CNN) have been successfully applied to many real-life problems. However, the huge memory cost of deep CNN models poses a great challenge of deploying them on memory-constrained devices (e.g., mobile phones). One popular way to reduce the memory cost of deep CNN model is to train bina...
main
Machine Learning
10.1609/aaai.v35i9.17002
35
9
8235-8242
official
2010.02778
title_snapshot
10.1609/aaai.v35i9.16963
Accurate and Robust Feature Importance Estimation under Distribution Shifts
https://ojs.aaai.org/index.php/AAAI/article/view/16963
https://ojs.aaai.org/index.php/AAAI/article/download/16963/16770
[ "Jayaraman J. Thiagarajan", "Vivek Narayanaswamy", "Rushil Anirudh", "Peer-Timo Bremer", "Andreas Spanias" ]
With increasing reliance on the outcomes of black-box models in critical applications, post-hoc explainability tools that do not require access to the model internals are often used to enable humans understand and trust these models. In particular, we focus on the class of methods that can reveal the influence of input...
main
Machine Learning
10.1609/aaai.v35i9.16963
35
9
7891-7898
official
2009.14454
title_snapshot
10.1609/aaai.v35i9.16964
Variance Penalized On-Policy and Off-Policy Actor-Critic
https://ojs.aaai.org/index.php/AAAI/article/view/16964
https://ojs.aaai.org/index.php/AAAI/article/download/16964/16771
[ "Arushi Jain", "Gandharv Patil", "Ayush Jain", "Khimya Khetarpal", "Doina Precup" ]
Reinforcement learning algorithms are typically geared towards optimizing the expected return of an agent. However, in many practical applications, low variance in the return is desired to ensure the reliability of an algorithm. In this paper, we propose on-policy and off-policy actor-critic algorithms that optimize a ...
main
Machine Learning
10.1609/aaai.v35i9.16964
35
9
7899-7907
official
2102.01985
title_snapshot
10.1609/aaai.v35i9.16965
Constructing a Fair Classifier with Generated Fair Data
https://ojs.aaai.org/index.php/AAAI/article/view/16965
https://ojs.aaai.org/index.php/AAAI/article/download/16965/16772
[ "Taeuk Jang", "Feng Zheng", "Xiaoqian Wang" ]
Fairness in machine learning is getting rising attention as it is directly related to real-world applications and social problems. Recent methods have been explored to alleviate the discrimination between certain demographic groups that are characterized by sensitive attributes (such as race, age, or gender). Some stud...
main
Machine Learning
10.1609/aaai.v35i9.16965
35
9
7908-7916
official
null
null
10.1609/aaai.v35i9.16966
Neural Utility Functions
https://ojs.aaai.org/index.php/AAAI/article/view/16966
https://ojs.aaai.org/index.php/AAAI/article/download/16966/16773
[ "Porter Jenkins", "Ahmad Farag", "J. Stockton Jenkins", "Huaxiu Yao", "Suhang Wang", "Zhenhui Li" ]
Current neural network architectures have no mechanism for explicitly reasoning about item trade-offs. Such trade-offs are important for popular tasks such as recommendation. The main idea of this work is to give neural networks inductive biases that are inspired by economic theories. To this end, we propose Neural Uti...
main
Machine Learning
10.1609/aaai.v35i9.16966
35
9
7917-7925
official
null
null
10.1609/aaai.v35i9.16967
IB-GAN: Disentangled Representation Learning with Information Bottleneck Generative Adversarial Networks
https://ojs.aaai.org/index.php/AAAI/article/view/16967
https://ojs.aaai.org/index.php/AAAI/article/download/16967/16774
[ "Insu Jeon", "Wonkwang Lee", "Myeongjang Pyeon", "Gunhee Kim" ]
We propose a new GAN-based unsupervised model for disentangled representation learning. The new model is discovered in an attempt to utilize the Information Bottleneck (IB) framework to the optimization of GAN, thereby named IB-GAN. The architecture of IB-GAN is partially similar to that of InfoGAN but has a critical d...
main
Machine Learning
10.1609/aaai.v35i9.16967
35
9
7926-7934
official
2510.20165
title_snapshot
10.1609/aaai.v35i9.16968
Active Bayesian Assessment of Black-Box Classifiers
https://ojs.aaai.org/index.php/AAAI/article/view/16968
https://ojs.aaai.org/index.php/AAAI/article/download/16968/16775
[ "Disi Ji", "Robert L. Logan", "Padhraic Smyth", "Mark Steyvers" ]
Recent advances in machine learning have led to increased deployment of black-box classifiers across a wide variety of applications. In many such situations there is a critical need to both reliably assess the performance of these pre-trained models and to perform this assessment in a label-efficient manner (given that...
main
Machine Learning
10.1609/aaai.v35i9.16968
35
9
7935-7944
official
2002.06532
title_judge
10.1609/aaai.v35i9.16969
Show, Attend and Distill: Knowledge Distillation via Attention-based Feature Matching
https://ojs.aaai.org/index.php/AAAI/article/view/16969
https://ojs.aaai.org/index.php/AAAI/article/download/16969/16776
[ "Mingi Ji", "Byeongho Heo", "Sungrae Park" ]
Knowledge distillation extracts general knowledge from a pretrained teacher network and provides guidance to a target student network. Most studies manually tie intermediate features of the teacher and student, and transfer knowledge through predefined links. However, manual selection often constructs ineffective links...
main
Machine Learning
10.1609/aaai.v35i9.16969
35
9
7945-7952
official
2102.02973
title_snapshot
10.1609/aaai.v35i9.16970
Dynamic Multi-Context Attention Networks for Citation Forecasting of Scientific Publications
https://ojs.aaai.org/index.php/AAAI/article/view/16970
https://ojs.aaai.org/index.php/AAAI/article/download/16970/16777
[ "Taoran Ji", "Nathan Self", "Kaiqun Fu", "Zhiqian Chen", "Naren Ramakrishnan", "Chang-Tien Lu" ]
Forecasting citations of scientific patents and publications is a crucial task for understanding the evolution and development of technological domains and for foresight into emerging technologies. By construing citations as a time series, the task can be cast into the domain of temporal point processes. Most existing ...
main
Machine Learning
10.1609/aaai.v35i9.16970
35
9
7953-7960
official
null
null
10.1609/aaai.v35i9.16971
Intrinsic Certified Robustness of Bagging against Data Poisoning Attacks
https://ojs.aaai.org/index.php/AAAI/article/view/16971
https://ojs.aaai.org/index.php/AAAI/article/download/16971/16778
[ "Jinyuan Jia", "Xiaoyu Cao", "Neil Zhenqiang Gong" ]
In a data poisoning attack, an attacker modifies, deletes, and/or inserts some training examples to corrupt the learnt machine learning model. Bootstrap Aggregating (bagging) is a well known ensemble learning method, which trains multiple base models on random subsamples of a training dataset using a base learning algo...
main
Machine Learning
10.1609/aaai.v35i9.16971
35
9
7961-7969
official
2008.04495
title_snapshot
10.1609/aaai.v35i9.16972
Clustering Ensemble Meets Low-rank Tensor Approximation
https://ojs.aaai.org/index.php/AAAI/article/view/16972
https://ojs.aaai.org/index.php/AAAI/article/download/16972/16779
[ "Yuheng Jia", "Hui Liu", "Junhui Hou", "Qingfu Zhang" ]
This paper explores the problem of clustering ensemble, which aims to combine multiple base clusterings to produce better performance than that of the individual one. The existing clustering ensemble methods generally construct a co-association matrix, which indicates the pairwise similarity between samples, as the wei...
main
Machine Learning
10.1609/aaai.v35i9.16972
35
9
7970-7978
official
2012.08916
title_snapshot
10.1609/aaai.v35i9.16973
Action Candidate Based Clipped Double Q-learning for Discrete and Continuous Action Tasks
https://ojs.aaai.org/index.php/AAAI/article/view/16973
https://ojs.aaai.org/index.php/AAAI/article/download/16973/16780
[ "Haobo Jiang", "Jin Xie", "Jian Yang" ]
Double Q-learning is a popular reinforcement learning algorithm in Markov decision process (MDP) problems. Clipped Double Q-learning, as an effective variant of Double Q-learning, employs the clipped double estimator to approximate the maximum expected action value. Due to the underestimation bias of the clipped double...
main
Machine Learning
10.1609/aaai.v35i9.16973
35
9
7979-7986
official
2105.00704
title_snapshot
10.1609/aaai.v35i9.16974
LightXML: Transformer with Dynamic Negative Sampling for High-Performance Extreme Multi-label Text Classification
https://ojs.aaai.org/index.php/AAAI/article/view/16974
https://ojs.aaai.org/index.php/AAAI/article/download/16974/16781
[ "Ting Jiang", "Deqing Wang", "Leilei Sun", "Huayi Yang", "Zhengyang Zhao", "Fuzhen Zhuang" ]
Extreme multi-label text classification(XMC) is a task for finding the most relevant labels from a large label set. Nowadays deep learning-based methods have shown significant success in XMC. However, the existing methods (e.g., AttentionXML and X-Transformer etc) still suffer from 1) combining several models to train ...
main
Machine Learning
10.1609/aaai.v35i9.16974
35
9
7987-7994
official
2101.03305
title_snapshot
10.1609/aaai.v35i9.16975
Temporal-Logic-Based Reward Shaping for Continuing Reinforcement Learning Tasks
https://ojs.aaai.org/index.php/AAAI/article/view/16975
https://ojs.aaai.org/index.php/AAAI/article/download/16975/16782
[ "Yuqian Jiang", "Suda Bharadwaj", "Bo Wu", "Rishi Shah", "Ufuk Topcu", "Peter Stone" ]
In continuing tasks, average-reward reinforcement learning may be a more appropriate problem formulation than the more common discounted reward formulation. As usual, learning an optimal policy in this setting typically requires a large amount of training experiences. Reward shaping is a common approach for incorporati...
main
Machine Learning
10.1609/aaai.v35i9.16975
35
9
7995-8003
official
2007.01498
title_snapshot
10.1609/aaai.v35i9.16976
Power up! Robust Graph Convolutional Network via Graph Powering
https://ojs.aaai.org/index.php/AAAI/article/view/16976
https://ojs.aaai.org/index.php/AAAI/article/download/16976/16783
[ "Ming Jin", "Heng Chang", "Wenwu Zhu", "Somayeh Sojoudi" ]
Graph convolutional networks (GCNs) are powerful tools for graph-structured data. However, they have been recently shown to be vulnerable to topological attacks. To enhance adversarial robustness, we go beyond spectral graph theory to robust graph theory. By challenging the classical graph Laplacian, we propose a new c...
main
Machine Learning
10.1609/aaai.v35i9.16976
35
9
8004-8012
official
1905.10029
title_snapshot