Curricula for Learning Robust Policies with Factored State Representations in Changing Environments
This work addresses the challenge of adapting reinforcement learning agents to unpredictable real-world environments, though it appears incremental as it builds on existing factored representation methods.
The paper tackled the problem of improving policy robustness in reinforcement learning for dynamic environments by exploring how curricula affect agents with factored state representations, and demonstrated that three simple curricula significantly enhance robustness.
Robust policies enable reinforcement learning agents to effectively adapt to and operate in unpredictable, dynamic, and ever-changing real-world environments. Factored representations, which break down complex state and action spaces into distinct components, can improve generalization and sample efficiency in policy learning. In this paper, we explore how the curriculum of an agent using a factored state representation affects the robustness of the learned policy. We experimentally demonstrate three simple curricula, such as varying only the variable of highest regret between episodes, that can significantly enhance policy robustness, offering practical insights for reinforcement learning in complex environments.