ARLBench: Flexible and Efficient Benchmarking for Hyperparameter Optimization in Reinforcement Learning

TL;DR

ARLBench is a new benchmark designed to facilitate efficient and flexible comparison of hyperparameter optimization methods in reinforcement learning, enabling broader research with limited computational resources.

Contribution

It introduces a scalable benchmark and dataset for HPO in RL, allowing diverse approaches to be evaluated efficiently across multiple domains.

Findings

Enables comparison of HPO methods with reduced compute

Provides a representative set of RL tasks for benchmarking

Supports research in AutoRL with accessible resources

Abstract

Hyperparameters are a critical factor in reliably training well-performing reinforcement learning (RL) agents. Unfortunately, developing and evaluating automated approaches for tuning such hyperparameters is both costly and time-consuming. As a result, such approaches are often only evaluated on a single domain or algorithm, making comparisons difficult and limiting insights into their generalizability. We propose ARLBench, a benchmark for hyperparameter optimization (HPO) in RL that allows comparisons of diverse HPO approaches while being highly efficient in evaluation. To enable research into HPO in RL, even in settings with low compute resources, we select a representative subset of HPO tasks spanning a variety of algorithm and environment combinations. This selection allows for generating a performance profile of an automated RL (AutoRL) method using only a fraction of the compute previously necessary, enabling a broader range of researchers to work on HPO in RL. With the extensive and large-scale dataset on hyperparameter landscapes that our selection is based on, ARLBench is an efficient, flexible, and future-oriented foundation for research on AutoRL. Both the benchmark and the dataset are available at https://github.com/automl/arlbench.