For directory, see ICML2020-Accepted List.pdf

In this pdf file, papers highlighted in YELLOW/ORANGE are in this repository.

PINK highlighted means interesting papers not available online.

List of papers in topics: (without replacement, so each paper in the most revelant topic)

• A Unified Theory of Decentralized SGD with Changing Topology and Local Updates
• Adascale_sgd_a_scale_invariant_algorithm_for_distributed_training-Original Pdf
• Improving the Sample and Communication Complexity for Decentralized Non-Convex Optimization - A Joint Gradient Estimation and Tracking Approach
• Is Local SGD Better than Minibatch SGD/ On Convergence-Diagnostic based Step Sizes for Stochastic Gradient Descent
• On the Noisy Gradient Descent that Generalizes as SGD
• The_Complexity_of_Finding_Stationary_Points_with_Stochastic_Gradient_Descent

• Min-Max Optimization without Gradients - Convergence and Applications to Adversarial ML
• Second-Order Provable Defenses against Adversarial Attacks
• Zeno++

Almost Tune-Free Variance Reduction

• Acceleration for Compressed Gradient Descent in Distributed Optimization
• Convergence of a Stochastic Gradient Method with Momentum for Non-Smooth Non-Convex Optimization
• Momentum Improves Normalized SGD
• Statistically Preconditioned Accelerated Gradient Method for Distributed Optimization
• Universal Average-Case Optimality of Polyak Momentum

• COMMUNICATION-EFFICIENT FEDERATED LEARNING WITH SKETCHING
• Federated Learning with Only Positive Labels
• From Local SGD to Local Fixed Point Methods for Federated Learning
• SCAFFOLD Stochastic Controlled Averaging for Federated Learning

• High-Dimensional Robust Mean Estimation via Gradient Descent
• Safe Screening Rules for L0-regression
• Spectral Graph Matching and Regularized Quadratic Relaxation I - The Gaussian Model
• WONDER Weighted One-shot Distributed Ridge Regression in High Dimensions