Safe Zeroth-Order Convex Optimization Using Quadratic Local Approximations

被引:0
|
作者
Guo, Baiwei [1 ]
Jiang, Yuning [2 ]
Kamgarpour, Maryam [3 ]
Ferrari-Trecate, Giancarlo [1 ]
机构
[1] Ecole Polytech Fed Lausanne, Inst Engn Mech, DECODE Grp, Lausanne, Switzerland
[2] Ecole Polytech Fed Lausanne, Predict Grp, IGM, Lausanne, Switzerland
[3] Ecole Polytech Fed Lausanne, Syst Control & Multiagent Optimizat Res Sycamore, IGM, Lausanne, Switzerland
基金
瑞士国家科学基金会;
关键词
SEARCH ALGORITHM;
D O I
10.23919/ECC57647.2023.10178374
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
We address black-box convex optimization problems, where the objective and constraint functions are not explicitly known but can be sampled within the feasible set. The challenge is thus to generate a sequence of feasible points converging towards an optimal solution. By leveraging the knowledge of the smoothness properties of the objective and constraint functions, we propose a novel zeroth-order method, SZO-QQ, that iteratively computes quadratic approximations of the constraint functions, constructs local feasible sets and optimizes over them. We prove convergence of the sequence of the objective values generated at each iteration to the minimum. Through experiments, we show that our method can achieve faster convergence compared with state-of-the-art zeroth-order approaches to convex optimization.
引用
收藏
页数:8
相关论文
共 50 条
  • [1] Safe Zeroth-Order Optimization Using Linear Programs
    Guo, Baiwei
    Wang, Yang
    Jiang, Yuning
    Kamgarpour, Maryam
    Ferrari-Trecate, Giancarlo
    [J]. 2023 62ND IEEE CONFERENCE ON DECISION AND CONTROL, CDC, 2023, : 556 - 561
  • [2] A Generic Approach for Accelerating Stochastic Zeroth-Order Convex Optimization
    Yu, Xiaotian
    King, Irwin
    Lyu, Michael R.
    Yang, Tianbao
    [J]. PROCEEDINGS OF THE TWENTY-SEVENTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2018, : 3040 - 3046
  • [3] A ZEROTH-ORDER PROXIMAL STOCHASTIC GRADIENT METHOD FOR WEAKLY CONVEX STOCHASTIC OPTIMIZATION
    Pougkakiotis, Spyridon
    Kalogerias, Dionysis
    [J]. SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2023, 45 (05): : A2679 - A2702
  • [4] Accelerated Zeroth-Order Algorithm for Stochastic Distributed Non-Convex Optimization
    Zhang, Shengjun
    Bailey, Colleen P.
    [J]. 2022 AMERICAN CONTROL CONFERENCE, ACC, 2022, : 4274 - 4279
  • [5] Escaping Saddle Points for Zeroth-order Non-convex Optimization using Estimated Gradient Descent
    Bai, Qinbo
    Agarwal, Mridul
    Aggarwal, Vaneet
    [J]. 2020 54TH ANNUAL CONFERENCE ON INFORMATION SCIENCES AND SYSTEMS (CISS), 2020, : 132 - 137
  • [6] Automatic controller tuning using a zeroth-order optimization algorithm
    Zalkind, Daniel S.
    Dall'Anese, Emiliano
    Pao, Lucy Y.
    [J]. WIND ENERGY SCIENCE, 2020, 5 (04) : 1579 - 1600
  • [7] Zeroth-order (Non)-Convex Stochastic Optimization via Conditional Gradient and Gradient Updates
    Balasubramanian, Krishnakumar
    Ghadimi, Saeed
    [J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 31 (NIPS 2018), 2018, 31
  • [8] UNBIASED GRADIENT SIMULATION FOR ZEROTH-ORDER OPTIMIZATION
    Chen, Guanting
    [J]. 2020 WINTER SIMULATION CONFERENCE (WSC), 2020, : 2947 - 2959
  • [9] Stochastic Zeroth-order Optimization in High Dimensions
    Wang, Yining
    Du, Simon S.
    Balakrishnan, Sivaraman
    Singh, Aarti
    [J]. INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 84, 2018, 84
  • [10] Zeroth-order optimization with orthogonal random directions
    David Kozak
    Cesare Molinari
    Lorenzo Rosasco
    Luis Tenorio
    Silvia Villa
    [J]. Mathematical Programming, 2023, 199 : 1179 - 1219