A Compression-Inspired Framework for Macro Discovery
This work addresses the challenge of sample efficiency in reinforcement learning for agents facing multiple related tasks, though it appears incremental as it builds on existing macro discovery methods.
The paper tackles the problem of enabling reinforcement learning agents to reuse knowledge from previous tasks to solve new related tasks more quickly, by proposing a framework that discovers and uses macros from compressed trajectories, resulting in faster learning of optimal policies in unseen MDPs.
In this paper we consider the problem of how a reinforcement learning agent tasked with solving a set of related Markov decision processes can use knowledge acquired early in its lifetime to improve its ability to more rapidly solve novel, but related, tasks. One way of exploiting this experience is by identifying recurrent patterns in trajectories obtained from well-performing policies. We propose a three-step framework in which an agent 1) generates a set of candidate open-loop macros by compressing trajectories drawn from near-optimal policies; 2) evaluates the value of each macro; and 3) selects a maximally diverse subset of macros that spans the space of policies typically required for solving the set of related tasks. Our experiments show that extending the original primitive action-set of the agent with the identified macros allows it to more rapidly learn an optimal policy in unseen, but similar MDPs.