evosax: JAX-based Evolution Strategies

TL;DR

evosax is a JAX-based library that enables scalable, hardware-accelerated evolutionary strategies, facilitating efficient black-box optimization on GPUs and TPUs with a flexible, modular API.

Contribution

The paper introduces evosax, a novel JAX-based library that implements 30 evolutionary algorithms optimized for modern hardware accelerators, enhancing scalability and performance.

Findings

Enables direct execution of algorithms on GPUs and TPUs.

Supports automatic vectorization and parallelization across devices.

Provides a modular, easy-to-use API for evolutionary optimization.

Abstract

The deep learning revolution has greatly been accelerated by the 'hardware lottery': Recent advances in modern hardware accelerators and compilers paved the way for large-scale batch gradient optimization. Evolutionary optimization, on the other hand, has mainly relied on CPU-parallelism, e.g. using Dask scheduling and distributed multi-host infrastructure. Here we argue that also modern evolutionary computation can significantly benefit from the massive computational throughput provided by GPUs and TPUs. In order to better harness these resources and to enable the next generation of black-box optimization algorithms, we release evosax: A JAX-based library of evolution strategies which allows researchers to leverage powerful function transformations such as just-in-time compilation, automatic vectorization and hardware parallelization. evosax implements 30 evolutionary optimization algorithms including finite-difference-based, estimation-of-distribution evolution strategies and various genetic algorithms. Every single algorithm can directly be executed on hardware accelerators and automatically vectorized or parallelized across devices using a single line of code. It is designed in a modular fashion and allows for flexible usage via a simple ask-evaluate-tell API. We thereby hope to facilitate a new wave of scalable evolutionary optimization algorithms.