A Consciousness-Inspired Planning Agent for Model-Based Reinforcement Learning
This work addresses generalization challenges in reinforcement learning for agents operating in complex, unseen environments, though it appears incremental as it builds on existing model-based methods with a novel attention mechanism.
The paper tackles the problem of poor generalization in model-based reinforcement learning by introducing a consciousness-inspired planning agent that dynamically attends to relevant state parts, resulting in improved out-of-distribution generalization performance in customized environments.
We present an end-to-end, model-based deep reinforcement learning agent which dynamically attends to relevant parts of its state during planning. The agent uses a bottleneck mechanism over a set-based representation to force the number of entities to which the agent attends at each planning step to be small. In experiments, we investigate the bottleneck mechanism with several sets of customized environments featuring different challenges. We consistently observe that the design allows the planning agents to generalize their learned task-solving abilities in compatible unseen environments by attending to the relevant objects, leading to better out-of-distribution generalization performance.