paper-with-me

홈 › Papers

GDA-AM: On the effectiveness of solving minimax optimization via Anderson Acceleration

2021-10-06 · Huan He, Shifan Zhao, Yuanzhe Xi, Joyce C Ho, Yousef Saad

Many modern machine learning algorithms such as generative adversarial networks (GANs) and adversarial training can be formulated as minimax optimization. Gradient descent ascent (GDA) is the most commonly used algorithm due to its simplicity. However, GDA can converge to non-optimal minimax points. We propose a new minimax optimization framework, GDA-AM, that views the GDAdynamics as a fixed-point iteration and solves it using Anderson Mixing to con-verge to the local minimax. It addresses the diverging issue of simultaneous GDAand accelerates the convergence of alternating GDA. We show theoretically that the algorithm can achieve global convergence for bilinear problems under mild conditions. We also empirically show that GDA-AMsolves a variety of minimax problems and improves GAN training on several datasets

📄 PDF Abstract BibTeX arXiv:2110.02457

Code (1)

hehuannb/gda-am 공식 구현

Similar Papers 제목 키워드 기반

GDA-AM: ON THE EFFECTIVENESS OF SOLVING MIN-IMAX OPTIMIZATION VIA ANDERSON MIXING

2021-09-29 · ICLR 2022 4 · Huan He, Shifan Zhao, Yuanzhe Xi, Joyce Ho 외

Many modern machine learning algorithms such as generative adversarial networks (GANs) and adversarial training can be formulated as minimax optimization.Gradient descent ascent (GDA) is the most commonly used algorithm …

Anderson acceleration for iteratively reweighted $\ell_1$ algorithm

2024-03-12 · Kexin Li

Iteratively reweighted L1 (IRL1) algorithm is a common algorithm for solving sparse optimization problems with nonconvex and nonsmooth regularization. The development of its acceleration algorithm, often employing Nester…

Anderson Acceleration of Proximal Gradient Methods

2019-10-18 · ICML 2020 1 · Vien V. Mai, Mikael Johansson

Anderson acceleration is a well-established and simple technique for speeding up fixed-point computations with countless applications. Previous studies of Anderson acceleration in optimization have only been able to prov…

Anderson Acceleration For Bioinformatics-Based Machine Learning

2023-02-01 · Sarwan Ali, Prakash Chourasia, Murray Patterson

Anderson acceleration (AA) is a well-known method for accelerating the convergence of iterative algorithms, with applications in various fields including deep learning and optimization. Despite its popularity in these ar…

Regularized Anderson Acceleration for Off-Policy Deep Reinforcement Learning

2019-09-07 · NeurIPS 2019 12 · Wenjie Shi, Shiji Song, Hui Wu, Ya-Chu Hsu 외

Model-free deep reinforcement learning (RL) algorithms have been widely used for a range of complex control tasks. However, slow convergence and sample inefficiency remain challenging problems in RL, especially when hand…

Deep Reinforcement LearningMuJoCoreinforcement-learningReinforcement Learning+1