Gradient Descent Ascent for Minimax Problems on Riemannian Manifolds

被引:5
|
作者
Huang, Feihu [1 ,2 ]
Gao, Shangqian [3 ]
机构
[1] Nanjing Univ Aeronaut & Astronaut, Coll Comp Sci & Technol, Nanjing 210016, Jiangsu, Peoples R China
[2] MIIT Key Lab Pattern Anal & Machine Intelligence, Nanjing 211106, Jiangsu, Peoples R China
[3] Univ Pittsburgh, Dept Elect & Comp Engn, Pittsburgh, PA 15260 USA
关键词
Manifolds; Optimization; Training; Machine learning; Complexity theory; Principal component analysis; Neural networks; Deep neural networks; minimax optimization; riemanian manifolds; robust optimization; stiefel manifold; MONOTONE VECTOR-FIELDS; ALGORITHM;
D O I
10.1109/TPAMI.2023.3234160
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In the paper, we study a class of useful minimax problems on Riemanian manifolds and propose a class of effective Riemanian gradient-based methods to solve these minimax problems. Specifically, we propose an effective Riemannian gradient descent ascent (RGDA) algorithm for the deterministic minimax optimization. Moreover, we prove that our RGDA has a sample complexity of O(kappa(2)epsilon(-2)) for finding an epsilon-stationary solution of the Geodesically-Nonconvex Strongly-Concave (GNSC) minimax problems, where kappa denotes the condition number. At the same time, we present an effective Riemannian stochastic gradient descent ascent (RSGDA) algorithm for the stochastic minimax optimization, which has a sample complexity of O(kappa(4)epsilon(-4)) for finding an epsilon-stationary solution. To further reduce the sample complexity, we propose an accelerated Riemannian stochastic gradient descent ascent (Acc-RSGDA) algorithm based on the momentum-based variance-reduced technique. We prove that our Acc-RSGDA algorithm achieves a lower sample complexity of (O) over tilde(kappa(4)epsilon(-3)) in searching for an epsilon-stationary solution of the GNSC minimax problems. Extensive experimental results on the robust distributional optimization and robust Deep Neural Networks (DNNs) training over Stiefel manifold demonstrate efficiency of our algorithms.
引用
收藏
页码:8466 / 8476
页数:11
相关论文
共 50 条
  • [21] Randomized Stochastic Gradient Descent Ascent
    Sebbouh, Othmane
    Cuturi, Marco
    Peyre, Gabriel
    INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 151, 2022, 151
  • [22] A Riemannian gradient ascent algorithm with applications to orthogonal approximation problems of symmetric tensors
    Sheng, Zhou
    Yang, Weiwei
    Wen, Jie
    APPLIED NUMERICAL MATHEMATICS, 2022, 182 : 235 - 247
  • [23] Ascent, descent and additive preserving problems
    Oudghiri, Mourad
    Souilah, Khalid
    STUDIA UNIVERSITATIS BABES-BOLYAI MATHEMATICA, 2019, 64 (04): : 565 - 580
  • [24] Coordinate Descent Without Coordinates: Tangent Subspace Descent on Riemannian Manifolds
    Gutman, David H.
    Ho-Nguyen, Nam
    MATHEMATICS OF OPERATIONS RESEARCH, 2023, 48 (01) : 127 - 159
  • [25] THE GENERALIZED CONDITIONAL GRADIENT METHOD FOR COMPOSITE MULTIOBJECTIVE OPTIMIZATION PROBLEMS ON RIEMANNIAN MANIFOLDS
    Li, Xiaobo
    Ge, Xiaochun
    Tu, Kai
    JOURNAL OF NONLINEAR AND VARIATIONAL ANALYSIS, 2023, 7 (05): : 839 - 857
  • [26] Sufficient Descent Riemannian Conjugate Gradient Methods
    Sakai, Hiroyuki
    Iiduka, Hideaki
    JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS, 2021, 190 (01) : 130 - 150
  • [27] Sufficient Descent Riemannian Conjugate Gradient Methods
    Hiroyuki Sakai
    Hideaki Iiduka
    Journal of Optimization Theory and Applications, 2021, 190 : 130 - 150
  • [28] Decentralized Riemannian Gradient Descent on the Stiefel Manifold
    Chen, Shixiang
    Garcia, Alfredo
    Hong, Mingyi
    Shahrampour, Shahin
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 139, 2021, 139
  • [29] Alternating Gradient Descent Ascent for Nonconvex Min-Max Problems in Robust Learning and GANs
    Lu, Songtao
    Singh, Rahul
    Chen, Xiangyi
    Chen, Yongxin
    Hong, Mingyi
    CONFERENCE RECORD OF THE 2019 FIFTY-THIRD ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS & COMPUTERS, 2019, : 680 - 684
  • [30] On Convergence of Gradient Descent Ascent: A Tight Local Analysis
    Li, Haochuan
    Farnia, Farzan
    Das, Subhro
    Jadbabaie, Ali
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 162, 2022,