A Hybrid Variance-Reduced Method for Decentralized Stochastic Non-Convex Optimization
Comments Accepted in ICML 2021
期刊&会议
ICML
International Conference on Machine Learning · 会议 · Machine Learning
Towards Certifying L-infinity Robustness using Neural Networks with L-inf-dist Neurons
Comments Appearing at International Conference on Machine Learning (ICML) 2021
Finding the Stochastic Shortest Path with Low Regret: The Adversarial Cost and Unknown Transition Case
Comments ICML 2021
SPADE: A Spectral Method for Black-Box Adversarial Robustness Evaluation
Comments The 2021 International Conference on Machine Learning (ICML)
Learning Noise Transition Matrix from Only Noisy Labels via Total Variation Regularization
Comments ICML 2021 camera-ready version
Information Obfuscation of Graph Neural Networks
Comments ICML 2021; Code is available at https://github.com/liaopeiyuan/GAL
Learning Randomly Perturbed Structured Predictors for Direct Loss Minimization
Comments Proceedings of the 38th International Conference on Machine Learning, 2021
The Heavy-Tail Phenomenon in SGD
Journal ref Published as a conference paper at International Conference on Machine Learning (ICML) 2021
Federated Continual Learning with Weighted Inter-client Transfer
Comments ICML 2021
SKIing on Simplices: Kernel Interpolation on the Permutohedral Lattice for Scalable Gaussian Processes
Comments International Conference on Machine Learning (ICML), 2021
CARTL: Cooperative Adversarially-Robust Transfer Learning
Comments Published as a conference paper at ICML 2021
Equivariant Networks for Pixelized Spheres
Comments Accepted to ICML 2021
Robust Representation Learning via Perceptual Similarity Metrics
Comments Accepted to ICML 2021
Skew Orthogonal Convolutions
Comments Accepted at ICML, 2021
Label Inference Attacks from Log-loss Scores
Comments Accepted at ICML 2021
EL-Attention: Memory Efficient Lossless Attention for Generation
Comments ICML 2021. Version 2: add pseudocode
Meta-Learning Bidirectional Update Rules
Comments ICML 2021, 17 pages
Acceleration via Fractal Learning Rate Schedules
Comments v2: revisions for ICML 2021
GRAD-MATCH: Gradient Matching based Data Subset Selection for Efficient Deep Model Training
Comments To appear in Proceedings of the 38 th International Conference on Machine Learning, PMLR 139, 2021
Sinkhorn Label Allocation: Semi-Supervised Classification via Annealed Self-Training
Comments ICML 2021 camera ready version
Globally-Robust Neural Networks
Comments Appearing in ICML 2021
Scaling Multi-Agent Reinforcement Learning with Selective Parameter Sharing
Comments To be published In Proceedings of the 38th International Conference on Machine Learning (ICML), 2021
CATE: Computation-aware Neural Architecture Encoding with Transformers
Comments ICML 2021 camera-ready. Code: https://github.com/MSU-MLSys-Lab/CATE
Locally Adaptive Label Smoothing for Predictive Churn
Comments ICML 2021
Grounding Language to Entities and Dynamics for Generalization in Reinforcement Learning
Comments Accepted to ICML 2021. Note author list and name changes from previous version
Locally Persistent Exploration in Continuous Control Tasks with Sparse Rewards
Comments To be published in ICML, 2021
A Policy Gradient Algorithm for Learning to Learn in Multiagent Reinforcement Learning
Comments Accepted to ICML 2021. Code at https://github.com/dkkim93/meta-mapg and Videos at https://sites.google.com/view/meta-mapg/home
Machine Unlearning for Random Forests
Comments 29 pages, 5 figures, 9 tables, and 3 algorithms. Accepted at ICML 2021
PAGE: A Simple and Optimal Probabilistic Gradient Estimator for Nonconvex Optimization
Comments 25 pages; accepted by ICML 2021 (long talk)
SUNRISE: A Simple Unified Framework for Ensemble Learning in Deep Reinforcement Learning
Comments ICML 2021 camera ready
