Contextualize Me -- The Case for Context in Reinforcement Learning
This work addresses the problem of poor generalization in reinforcement learning for researchers and practitioners, though it is incremental as it builds on existing cRL concepts with new benchmarks.
The paper tackles the brittleness of reinforcement learning algorithms to environmental changes by proposing Contextual Reinforcement Learning (cRL) as a framework for modeling such changes, and introduces CARL, a benchmark library that shows simple RL environments become challenging in contextual settings, with naive solutions failing to generalize across complex context spaces.
While Reinforcement Learning ( RL) has made great strides towards solving increasingly complicated problems, many algorithms are still brittle to even slight environmental changes. Contextual Reinforcement Learning (cRL) provides a framework to model such changes in a principled manner, thereby enabling flexible, precise and interpretable task specification and generation. Our goal is to show how the framework of cRL contributes to improving zero-shot generalization in RL through meaningful benchmarks and structured reasoning about generalization tasks. We confirm the insight that optimal behavior in cRL requires context information, as in other related areas of partial observability. To empirically validate this in the cRL framework, we provide various context-extended versions of common RL environments. They are part of the first benchmark library, CARL, designed for generalization based on cRL extensions of popular benchmarks, which we propose as a testbed to further study general agents. We show that in the contextual setting, even simple RL environments become challenging - and that naive solutions are not enough to generalize across complex context spaces.