Dissipativity Theory for Accelerating Stochastic Variance Reduction: A Unified Analysis of SVRG and Katyusha Using Semidefinite Programs

被引:0
|
作者
Hu, Bin [1 ]
Wright, Stephen [1 ]
Lessard, Laurent [1 ]
机构
[1] Univ Wisconsin Madison, Madison, WI 53706 USA
基金
美国国家科学基金会;
关键词
DYNAMICAL-SYSTEMS; OPTIMIZATION;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Techniques for reducing the variance of gradient estimates used in stochastic programming algorithms for convex finite-sum problems have received a great deal of attention in recent years. By leveraging dissipativity theory from control, we provide a new perspective on two important variance-reduction algorithms: SVRG and its direct accelerated variant Katyusha. Our perspective provides a physically intuitive understanding of the behavior of SVRG-like methods via a principle of energy conservation. The tools discussed here allow us to automate the convergence analysis of SVRG-like methods by capturing their essential properties in small semidefinite programs amenable to standard analysis and computational techniques. Our approach recovers existing convergence results for SVRG and Katyusha and generalizes the theory to alternative parameter choices. We also discuss how our approach complements the linear coupling technique. Our combination of perspectives leads to a better understanding of accelerated variance-reduced stochastic methods for finite-sum problems.
引用
收藏
页数:10
相关论文
共 6 条
  • [1] Analysis of biased stochastic gradient descent using sequential semidefinite programs
    Hu, Bin
    Seiler, Peter
    Lessard, Laurent
    MATHEMATICAL PROGRAMMING, 2021, 187 (1-2) : 383 - 408
  • [2] Analysis of biased stochastic gradient descent using sequential semidefinite programs
    Bin Hu
    Peter Seiler
    Laurent Lessard
    Mathematical Programming, 2021, 187 : 383 - 408
  • [3] Accelerating stochastic sequential quadratic programming for equality constrained optimization using predictive variance reduction
    Berahas, Albert S.
    Shi, Jiahao
    Yi, Zihong
    Zhou, Baoyu
    COMPUTATIONAL OPTIMIZATION AND APPLICATIONS, 2023, 86 (01) : 79 - 116
  • [4] Accelerating stochastic sequential quadratic programming for equality constrained optimization using predictive variance reduction
    Albert S. Berahas
    Jiahao Shi
    Zihong Yi
    Baoyu Zhou
    Computational Optimization and Applications, 2023, 86 : 79 - 116
  • [5] Fast variance reduction for steady-state simulation and sensitivity analysis of stochastic chemical systems using shadow function estimators
    Milias-Argeitis, Andreas
    Lygeros, John
    Khammash, Mustafa
    JOURNAL OF CHEMICAL PHYSICS, 2014, 141 (02):
  • [6] Analysis of a detailed multi-stage model of stochastic gene expression using queueing theory and model reduction
    Ma, Muhan
    Szavits-Nossan, Juraj
    Singh, Abhyudai
    Grima, Ramon
    MATHEMATICAL BIOSCIENCES, 2024, 373