A Primer on Zeroth-Order Optimization in Signal Processing and Machine Learning

TL;DR

This paper reviews zeroth-order optimization, a gradient-free method used in signal processing and machine learning, highlighting its principles, recent advances, and diverse applications like robustness evaluation and sensor management.

Contribution

It provides a comprehensive overview of ZO optimization, emphasizing intuition, principles, convergence analysis, and practical applications in various fields.

Findings

Demonstrates ZO optimization's effectiveness in black-box model explanations

Highlights recent convergence analysis advances

Showcases applications in sensor management and robustness evaluation

Abstract

Zeroth-order (ZO) optimization is a subset of gradient-free optimization that emerges in many signal processing and machine learning 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 solution update. In this paper, 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 models, and efficient online sensor management.