Results 11 to 20 of about 32,351,759 (213)
Tight Analysis of Extra-gradient and Optimistic Gradient Methods For Nonconvex Minimax Problems [PDF]
Neural Information Processing Systems, 2022 Despite the established convergence theory of Optimistic Gradient Descent Ascent (OGDA) and Extragradient (EG) methods for the convex-concave minimax problems, little is known about the theoretical guarantees of these methods in nonconvex settings.
Pouria Mahdavinia +3 more
semanticscholar +1 more source
Efficient Dataset Distillation via Minimax Diffusion [PDF]
Computer Vision and Pattern Recognition, 2023 Dataset distillation reduces the storage and computational consumption of training a network by generating a small surrogate dataset that encapsulates rich information of the original large-scale one.
Jianyang Gu +6 more
semanticscholar +1 more source
Escaping limit cycles: Global convergence for constrained nonconvex-nonconcave minimax problems [PDF]
International Conference on Learning Representations, 2023 This paper introduces a new extragradient-type algorithm for a class of nonconvex-nonconcave minimax problems. It is well-known that finding a local solution for general minimax problems is computationally intractable.
T. Pethick +4 more
semanticscholar +1 more source
Decentralized Riemannian Algorithm for Nonconvex Minimax Problems [PDF]
AAAI Conference on Artificial Intelligence, 2023 The minimax optimization over Riemannian manifolds (possibly nonconvex constraints) has been actively applied to solve many problems, such as robust dimensionality reduction and deep neural networks with orthogonal weights (Stiefel manifold).
Xidong Wu, Zhengmian Hu, Heng Huang
semanticscholar +1 more source
FEDNEST: Federated Bilevel, Minimax, and Compositional Optimization [PDF]
International Conference on Machine Learning, 2022 Standard federated optimization methods successfully apply to stochastic problems with single-level structure. However, many contemporary ML problems -- including adversarial robustness, hyperparameter tuning, and actor-critic -- fall under nested ...
Davoud Ataee Tarzanagh +3 more
semanticscholar +1 more source
Minimax Demographic Group Fairness in Federated Learning [PDF]
Conference on Fairness, Accountability and Transparency, 2022 Federated learning is an increasingly popular paradigm that enables a large number of entities to collaboratively learn better models. In this work, we study minimax group fairness in federated learning scenarios where different participating entities ...
Afroditi Papadaki +4 more
semanticscholar +1 more source
Decentralized Stochastic Gradient Descent Ascent for Finite-Sum Minimax Problems [PDF]
arXiv.org, 2022 Minimax optimization problems have attracted significant attention in recent years due to their widespread application in numerous machine learning models. To solve the minimax problem, a wide variety of stochastic optimization methods have been proposed.
Hongchang Gao
semanticscholar +1 more source
Near-Minimax Optimal Estimation With Shallow ReLU Neural Networks [PDF]
IEEE Transactions on Information Theory, 2021 We study the problem of estimating an unknown function from noisy data using shallow ReLU neural networks. The estimators we study minimize the sum of squared data-fitting errors plus a regularization term proportional to the squared Euclidean norm of ...
Rahul Parhi, R. Nowak
semanticscholar +1 more source
GAN‐LSTM‐3D: An efficient method for lung tumour 3D reconstruction enhanced by attention‐based LSTM
CAAI Transactions on Intelligence Technology, EarlyView., 2023 Abstract Three‐dimensional (3D) image reconstruction of tumours can visualise their structures with precision and high resolution. In this article, GAN‐LSTM‐3D method is proposed for 3D reconstruction of lung cancer tumours from 2D CT images. Our method consists of three phases: lung segmentation, tumour segmentation, and tumour 3D reconstruction. Lung
Lu Hong +12 more
wiley +1 more source
Existence and Minimax Theorems for Adversarial Surrogate Risks in Binary Classification [PDF]
Journal of machine learning research, 2022 Adversarial training is one of the most popular methods for training methods robust to adversarial attacks, however, it is not well-understood from a theoretical perspective.
Natalie Frank, Jonathan Niles-Weed
semanticscholar +1 more source