Simple random search provides a competitive approach to reinforcement learning

TL;DR

This paper demonstrates that simple random search can be a competitive and efficient method for reinforcement learning, matching state-of-the-art results and outperforming other model-free approaches in continuous control tasks.

Contribution

Introducing a random search method for training linear policies that achieves competitive sample efficiency and computational performance in continuous control and classical control problems.

Findings

Random search matches state-of-the-art sample efficiency on MuJoCo tasks.

Our method finds a nearly optimal controller for the Linear Quadratic Regulator.

Random search is at least 15 times more computationally efficient than competing methods.

Abstract

A common belief in model-free reinforcement learning is that methods based on random search in the parameter space of policies exhibit significantly worse sample complexity than those that explore the space of actions. We dispel such beliefs by introducing a random search method for training static, linear policies for continuous control problems, matching state-of-the-art sample efficiency on the benchmark MuJoCo locomotion tasks. Our method also finds a nearly optimal controller for a challenging instance of the Linear Quadratic Regulator, a classical problem in control theory, when the dynamics are not known. Computationally, our random search algorithm is at least 15 times more efficient than the fastest competing model-free methods on these benchmarks. We take advantage of this computational efficiency to evaluate the performance of our method over hundreds of random seeds and many different hyperparameter configurations for each benchmark task. Our simulations highlight a high variability in performance in these benchmark tasks, suggesting that commonly used estimations of sample efficiency do not adequately evaluate the performance of RL algorithms.