Algorithms of Robust Stochastic Optimization Based on Mirror Descent Method

被引：21

作者：

Nazin, A., V ^{[1
]}

Nemirovsky, A. S. ^{[2
]}

Tsybakov, A. B. ^{[3
]}

Juditsky, A. B. ^{[4
]}

机构：

[1] Russian Acad Sci, Trapeznikov Inst Control Sci, Moscow, Russia

[2] Georgia Inst Technol, Atlanta, GA 30332 USA

[3] ENSAE, CREST, Palaiseau, France

[4] Univ Grenoble Alpes, Grenoble, France

来源：

AUTOMATION AND REMOTE CONTROL | 2019年 / 80卷 / 09期

基金：

俄罗斯科学基金会;

关键词：

robust iterative algorithms; stochastic optimization algorithms; convex composite stochastic optimization; mirror descent method; robust confidence sets; APPROXIMATION ALGORITHMS; CONVERGENCE;

D O I：

10.1134/S0005117919090042

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 propose an approach to the construction of robust non-Euclidean iterative algorithms by convex composite stochastic optimization based on truncation of stochastic gradients. For such algorithms, we establish sub-Gaussian confidence bounds under weak assumptions about the tails of the noise distribution in convex and strongly convex settings. Robust estimates of the accuracy of general stochastic algorithms are also proposed.

引用

页码：1607 / 1627

页数：21

共 50 条

[11] Stochastic incremental mirror descent algorithms with Nesterov smoothing
Bitterlich, Sandy
Grad, Sorin-Mihai
[J]. NUMERICAL ALGORITHMS, 2024, 95 (01) : 351 - 382
[12] A CHARACTERIZATION OF STOCHASTIC MIRROR DESCENT ALGORITHMS AND THEIR CONVERGENCE PROPERTIES
Azizan, Navid
Hassibi, Babak
[J]. 2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2019, : 5167 - 5171
[13] Stochastic incremental mirror descent algorithms with Nesterov smoothing
Sandy Bitterlich
Sorin-Mihai Grad
[J]. Numerical Algorithms, 2024, 95 : 351 - 382
[14] Gossip-based distributed stochastic mirror descent for constrained optimization
Fang, Xianju
Zhang, Baoyong
Yuan, Deming
[J]. NEURAL NETWORKS, 2024, 175
[15] STOCHASTIC BLOCK MIRROR DESCENT METHODS FOR NONSMOOTH AND STOCHASTIC OPTIMIZATION
Dang, Cong D.
Lan, Guanghui
[J]. SIAM JOURNAL ON OPTIMIZATION, 2015, 25 (02) : 856 - 881
[16] Primal-Dual Mirror Descent Method for Constraint Stochastic Optimization Problems
Bayandina, A. S.
Gasnikov, A. V.
Gasnikova, E. V.
Matsievskii, S. V.
[J]. COMPUTATIONAL MATHEMATICS AND MATHEMATICAL PHYSICS, 2018, 58 (11) : 1728 - 1736
[17] Stochastic Mirror Descent in Variationally Coherent Optimization Problems
Zhou, Zhengyuan
Mertikopoulos, Panayotis
Bambos, Nicholas
Boyd, Stephen
Glynn, Peter
[J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 30 (NIPS 2017), 2017, 30
[18] Validation analysis of mirror descent stochastic approximation method
Guanghui Lan
Arkadi Nemirovski
Alexander Shapiro
[J]. Mathematical Programming, 2012, 134 : 425 - 458
[19] Validation analysis of mirror descent stochastic approximation method
Lan, Guanghui
Nemirovski, Arkadi
Shapiro, Alexander
[J]. MATHEMATICAL PROGRAMMING, 2012, 134 (02) : 425 - 458
[20] Optimal distributed stochastic mirror descent for strongly convex optimization
Yuan, Deming
Hong, Yiguang
Ho, Daniel W. C.
Jiang, Guoping
[J]. AUTOMATICA, 2018, 90 : 196 - 203

← 1 2 3 4 5 →