A Primer on Zeroth-Order Optimization in Signal Processing and Machine Learning: Principals, Recent Advances, and Applications

Sijia Liu, Pin Yu Chen, Bhavya Kailkhura, Gaoyuan Zhang, Alfred O. Hero, Pramod K. Varshney

Research output: Contribution to journalArticle

Abstract

Zeroth-order (ZO) optimization is a subset of gradient-free optimization that emerges in many signal processing and machine learning (ML) applications. It is used for solving optimization problems similarly to gradient-based methods. However, it does not require the gradient, using only function evaluations. Specifically, ZO optimization iteratively performs three major steps: gradient estimation, descent direction computation, and the solution update. In this article, we provide a comprehensive review of ZO optimization, with an emphasis on showing the underlying intuition, optimization principles, and recent advances in convergence analysis. Moreover, we demonstrate promising applications of ZO optimization, such as evaluating robustness and generating explanations from black-box deep learning (DL) models and efficient online sensor management.

Original languageEnglish (US)
Article number9186148
Pages (from-to)43-54
Number of pages12
JournalIEEE Signal Processing Magazine
Volume37
Issue number5
DOIs
StatePublished - Sep 2020

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Applied Mathematics

Fingerprint Dive into the research topics of 'A Primer on Zeroth-Order Optimization in Signal Processing and Machine Learning: Principals, Recent Advances, and Applications'. Together they form a unique fingerprint.

Cite this