Reinforcement Learning for Improving Agent Design
This work addresses the challenge of suboptimal agent design in RL, potentially benefiting robotics and assisted-design applications, though it appears incremental as it builds on existing frameworks.
The paper tackles the problem of optimizing an agent's physical design for reinforcement learning tasks, proposing a method to jointly learn the agent's structure and policy, resulting in improved task performance and facilitated policy learning.
In many reinforcement learning tasks, the goal is to learn a policy to manipulate an agent, whose design is fixed, to maximize some notion of cumulative reward. The design of the agent's physical structure is rarely optimized for the task at hand. In this work, we explore the possibility of learning a version of the agent's design that is better suited for its task, jointly with the policy. We propose an alteration to the popular OpenAI Gym framework, where we parameterize parts of an environment, and allow an agent to jointly learn to modify these environment parameters along with its policy. We demonstrate that an agent can learn a better structure of its body that is not only better suited for the task, but also facilitates policy learning. Joint learning of policy and structure may even uncover design principles that are useful for assisted-design applications. Videos of results at https://designrl.github.io/