Gradient-Based Adaptive Stochastic Search for Non-Differentiable Optimization

被引:38
|
作者
Zhou, Enlu [1 ]
Hu, Jiaqiao [2 ]
机构
[1] Georgia Inst Technol, H Milton Stewart Sch Ind & Syst Engn, Atlanta, GA 30332 USA
[2] SUNY Stony Brook, Dept Appl Math & Stat, Stony Brook, NY 11794 USA
基金
美国国家科学基金会;
关键词
Stochastic approximation; stochastic search; black-box optimization; ALGORITHM;
D O I
10.1109/TAC.2014.2310052
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In this paper, we propose a stochastic search algorithm for solving general optimization problems with little structure. The algorithm iteratively finds high quality solutions by randomly sampling candidate solutions from a parameterized distribution model over the solution space. The basic idea is to convert the original (possibly non-differentiable) problem into a differentiable optimization problem on the parameter space of the parameterized sampling distribution, and then use a direct gradient search method to find improved sampling distributions. Thus, the algorithm combines the robustness feature of stochastic search from considering a population of candidate solutions with the relative fast convergence speed of classical gradient methods by exploiting local differentiable structures. We analyze the convergence and converge rate properties of the proposed algorithm, and carry out numerical study to illustrate its performance.
引用
收藏
页码:1818 / 1832
页数:15
相关论文
共 50 条
  • [1] Discrete optimization via gradient-based adaptive stochastic search methods
    Chen, Xi
    Zhou, Enlu
    Hu, Jiaqiao
    [J]. IISE TRANSACTIONS, 2018, 50 (09) : 789 - 805
  • [2] COMBINING GRADIENT-BASED OPTIMIZATION WITH STOCHASTIC SEARCH
    Zhou, Enlu
    Hu, Jiaqiao
    [J]. 2012 WINTER SIMULATION CONFERENCE (WSC), 2012,
  • [3] Transforming a Non-Differentiable Rasterizer into a Differentiable One with Stochastic Gradient Estimation
    Deliot, Thomas
    Heitz, Eric
    Belcour, Laurent
    [J]. PROCEEDINGS OF THE ACM ON COMPUTER GRAPHICS AND INTERACTIVE TECHNIQUES, 2024, 7 (01)
  • [4] Gradient-Based Adaptive Stochastic Search for Simulation Optimization Over Continuous Space
    Zhou, Enlu
    Bhatnagar, Shalabh
    [J]. INFORMS JOURNAL ON COMPUTING, 2018, 30 (01) : 154 - 167
  • [5] Gradient-Based Markov Chain Monte Carlo for Bayesian Inference With Non-differentiable Priors
    Goldman, Jacob Vorstrup
    Sell, Torben
    Singh, Sumeetpal Sidhu
    [J]. JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2022, 117 (540) : 2182 - 2193
  • [6] SIMULATION OPTIMIZATION VIA GRADIENT-BASED STOCHASTIC SEARCH
    Zhou, Enlu
    Bhatnagar, Shalabh
    Chen, Xi
    [J]. PROCEEDINGS OF THE 2014 WINTER SIMULATION CONFERENCE (WSC), 2014, : 3869 - 3879
  • [7] Sub-gradient based projection neural networks for non-differentiable optimization problems
    Li, Guo-Cheng
    Dong, Zhi-Ling
    [J]. PROCEEDINGS OF 2008 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS, VOLS 1-7, 2008, : 835 - 839
  • [8] Reparameterization Gradient for Non-differentiable Models
    Lee, Wonyeol
    Yu, Hangyeol
    Yang, Hongseok
    [J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 31 (NIPS 2018), 2018, 31
  • [9] Non-parametric Smoothing for Gradient Methods in Non-differentiable Optimization Problems
    Chakraborty, Arindam
    Roy, Arunjyoti Sinha
    Dasgupta, Bhaskar
    [J]. 2016 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC), 2016, : 3759 - 3764
  • [10] ON THE ADAPTIVITY OF STOCHASTIC GRADIENT-BASED OPTIMIZATION
    Lei, Lihua
    Jordan, Michael I.
    [J]. SIAM JOURNAL ON OPTIMIZATION, 2020, 30 (02) : 1473 - 1500