Emergence of Locomotion Behaviours in Rich Environments

TL;DR

This paper demonstrates that training reinforcement learning agents in diverse, rich environments naturally leads to the emergence of complex locomotion behaviors like running, jumping, and turning without explicit reward shaping.

Contribution

It introduces a scalable policy gradient method and shows that environmental diversity alone can promote complex behavior emergence in RL agents.

Findings

Agents learn diverse locomotion skills in challenging terrains

Rich environments reduce the need for explicit reward engineering

Behaviors emerge robustly across multiple tasks

Abstract

The reinforcement learning paradigm allows, in principle, for complex behaviours to be learned directly from simple reward signals. In practice, however, it is common to carefully hand-design the reward function to encourage a particular solution, or to derive it from demonstration data. In this paper explore how a rich environment can help to promote the learning of complex behavior. Specifically, we train agents in diverse environmental contexts, and find that this encourages the emergence of robust behaviours that perform well across a suite of tasks. We demonstrate this principle for locomotion -- behaviours that are known for their sensitivity to the choice of reward. We train several simulated bodies on a diverse set of challenging terrains and obstacles, using a simple reward function based on forward progress. Using a novel scalable variant of policy gradient reinforcement learning, our agents learn to run, jump, crouch and turn as required by the environment without explicit reward-based guidance. A visual depiction of highlights of the learned behavior can be viewed following https://youtu.be/hx_bgoTF7bs .