Eigenoption Discovery through the Deep Successor Representation
This work addresses the problem of hierarchical task decomposition for reinforcement learning agents, offering an incremental improvement by adapting existing methods to more complex settings.
The paper tackles the challenge of autonomously learning effective options in reinforcement learning by extending eigenoption discovery to stochastic transitions and raw pixel inputs, resulting in an algorithm that demonstrates potential in Atari 2600 games.
Options in reinforcement learning allow agents to hierarchically decompose a task into subtasks, having the potential to speed up learning and planning. However, autonomously learning effective sets of options is still a major challenge in the field. In this paper we focus on the recently introduced idea of using representation learning methods to guide the option discovery process. Specifically, we look at eigenoptions, options obtained from representations that encode diffusive information flow in the environment. We extend the existing algorithms for eigenoption discovery to settings with stochastic transitions and in which handcrafted features are not available. We propose an algorithm that discovers eigenoptions while learning non-linear state representations from raw pixels. It exploits recent successes in the deep reinforcement learning literature and the equivalence between proto-value functions and the successor representation. We use traditional tabular domains to provide intuition about our approach and Atari 2600 games to demonstrate its potential.