Abstract In this paper, we consider stochastic weakly convex optimization problems, however without the existence of a stochastic subgradient oracle. We present a derivative free algorithm that uses a two point approximation for computing a gradient estimate of the smoothed function. We prove convergence at a similar rate as state of the art methods, however with a larger constant, and report some numerical results showing the effectiveness of the approach.

V. Kungurtsev

F. Rinaldi

Computational Optimization and Applications

COMPUTATIONAL OPTIMIZATION AND APPLICATIONS 是一本同行评审期刊，致力于及时发表关于优化计算算法和建模技术的分析和开发的研究和教程论文。对于一般类别的优化问题或更具体的应用问题的算法是令人感兴趣的。将考虑随机算法和确定性算法。特别欢迎能够提供理论分析和精心设计的计算实验的论文。感兴趣的主题包括但不限于以下内容：大规模优化、无约束优化、线性规划、二次规划互补问题和变分不等式,约束优化,不可微优化,整数规划,组合优化,随机优化,多目标优化,网络优化,复杂性理论,近似和误差分析,参数规划和敏感性分析,并行计算,分布式计算,和向量处理,软件,基准,数值实验和比较，用于优化的建模语言和系统，自动微分，在工程、金融、优化控制、优化设计、运筹学、运输、经济学、通信、制造和管理科学中的应用。

Computational Optimization and Applications is a peer-reviewed journal dedicated to the timely publication of research and tutorial papers on the analysis and development of optimization computational algorithms and modeling techniques. The journal is interested in algorithms for general optimization problems as well as more specific application problems, including both stochastic and deterministic algorithms. Papers that provide theoretical analysis and well-designed computational experiments are particularly welcome. Topics of interest include, but are not limited to: large-scale optimization, unconstrained optimization, linear programming, quadratic programming complementarity problems, variational inequalities, constrained optimization, nonsmooth optimization, integer programming, combinatorial optimization, stochastic optimization, multi-objective optimization, network optimization, complexity theory, approximation and error analysis, parametric programming and sensitivity analysis, parallel computing, distributed computing, vector processing, software, benchmarks, numerical experiments and comparisons, modeling languages and systems for optimization, automatic differentiation, and applications in engineering, finance, optimization control, optimization design, operations research, transportation, economics, communications, manufacturing, and management science.

《Computational Optimization and Applications》是一本经过同行评审的期刊，专注于及时发表关于优化计算算法和建模技术的研究和教程论文。期刊欢迎涉及一般优化问题或特定应用问题的算法研究，涵盖随机算法和确定性算法。特别鼓励提供理论分析和精心设计的计算实验的论文。感兴趣的主题包括但不限于：大规模优化、无约束优化、线性规划、二次规划互补问题、变分不等式、约束优化、不可微优化、整数规划、组合优化、随机优化、多目标优化、网络优化、复杂性理论、近似和误差分析、参数规划和敏感性分析、并行计算、分布式计算、向量处理、软件、基准、数值实验和比较，以及用于优化的建模语言和系统、自动微分等在工程、金融、优化控制、优化设计、运筹学、运输、经济学、通信、制造和管理科学中的应用。

A zeroth order method for stochastic weakly convex optimization

A derivative free algorithm that uses a two point approximation for computing a gradient estimate of the smoothed function is presented, which proves convergence at a similar rate as state of the art methods, however with a larger constant.

Category	Quartile
数学	2
数学, 应用数学	2
数学, 运筹学与管理科学	3

A zeroth order method for stochastic weakly convex optimization

ivySCI AI Smartly Parses PDF, Answers Researchers' Questions, and Helps You Understand Papers in Seconds

Journal Info

Category	Quartile
OPERATIONS RESEARCH & MANAGEMENT SCIENCE	3