Offline reinforcement learning with realizability and single-policy concentrability

W Zhan, B Huang, A Huang… - … on Learning Theory, 2022 - proceedings.mlr.press
Sample-efficiency guarantees for offline reinforcement learning (RL) often rely on strong
assumptions on both the function classes (eg, Bellman-completeness) and the data …

The complexity of constrained min-max optimization

C Daskalakis, S Skoulakis, M Zampetakis - Proceedings of the 53rd …, 2021 - dl.acm.org
Despite its important applications in Machine Learning, min-max optimization of objective
functions that are nonconvex-nonconcave remains elusive. Not only are there no known first …

Distributionally robust federated averaging

Y Deng, MM Kamani… - Advances in neural …, 2020 - proceedings.neurips.cc
In this paper, we study communication efficient distributed algorithms for distributionally
robust federated learning via periodic averaging with adaptive sampling. In contrast to …

Accelerated Algorithms for Smooth Convex-Concave Minimax Problems with O (1/k^ 2) Rate on Squared Gradient Norm

TH Yoon, EK Ryu - International Conference on Machine …, 2021 - proceedings.mlr.press
In this work, we study the computational complexity of reducing the squared gradient
magnitude for smooth minimax optimization problems. First, we present algorithms with …

Efficient methods for structured nonconvex-nonconcave min-max optimization

J Diakonikolas, C Daskalakis… - … Conference on Artificial …, 2021 - proceedings.mlr.press
The use of min-max optimization in the adversarial training of deep neural network
classifiers, and the training of generative adversarial networks has motivated the study of …

Sublinear convergence rates of extragradient-type methods: A survey on classical and recent developments

Q Tran-Dinh - arxiv preprint arxiv:2303.17192, 2023 - arxiv.org
The extragradient (EG), introduced by GM Korpelevich in 1976, is a well-known method to
approximate solutions of saddle-point problems and their extensions such as variational …

Global convergence and variance reduction for a class of nonconvex-nonconcave minimax problems

J Yang, N Kiyavash, N He - Advances in Neural Information …, 2020 - proceedings.neurips.cc
Nonconvex minimax problems appear frequently in emerging machine learning
applications, such as generative adversarial networks and adversarial learning. Simple …

A single-loop smoothed gradient descent-ascent algorithm for nonconvex-concave min-max problems

J Zhang, P **ao, R Sun, Z Luo - Advances in neural …, 2020 - proceedings.neurips.cc
Nonconvex-concave min-max problem arises in many machine learning applications
including minimizing a pointwise maximum of a set of nonconvex functions and robust …

Faster single-loop algorithms for minimax optimization without strong concavity

J Yang, A Orvieto, A Lucchi… - … Conference on Artificial …, 2022 - proceedings.mlr.press
Gradient descent ascent (GDA), the simplest single-loop algorithm for nonconvex minimax
optimization, is widely used in practical applications such as generative adversarial …

Federated minimax optimization: Improved convergence analyses and algorithms

P Sharma, R Panda, G Joshi… - … on Machine Learning, 2022 - proceedings.mlr.press
In this paper, we consider nonconvex minimax optimization, which is gaining prominence in
many modern machine learning applications, such as GANs. Large-scale edge-based …