Differential Evolution for Neural Architecture Search

TL;DR

This paper introduces differential evolution as a new search strategy for neural architecture search, demonstrating its effectiveness and robustness across multiple NAS benchmarks using full evaluation methods.

Contribution

It applies differential evolution to NAS, providing a simple yet effective alternative to existing search strategies and showing improved results on standard benchmarks.

Findings

Differential evolution outperforms regularized evolution and Bayesian optimization.

It yields more robust results across 13 NAS benchmarks.

The approach is effective with full evaluation strategies.

Abstract

Neural architecture search (NAS) methods rely on a search strategy for deciding which architectures to evaluate next and a performance estimation strategy for assessing their performance (e.g., using full evaluations, multi-fidelity evaluations, or the one-shot model). In this paper, we focus on the search strategy. We introduce the simple yet powerful evolutionary algorithm of differential evolution to the NAS community. Using the simplest performance evaluation strategy of full evaluations, we comprehensively compare this search strategy to regularized evolution and Bayesian optimization and demonstrate that it yields improved and more robust results for 13 tabular NAS benchmarks based on NAS-Bench-101, NAS-Bench-1Shot1, NAS-Bench-201 and NAS-HPO bench.