A Consciousness-Inspired Planning Agent for Model-Based Reinforcement Learning
arXiv:2106.02097
Abstract
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.
arXiv commentNeurIPS camera-ready version
References in corpus (6)
- Recurrent World Models Facilitate Policy Evolution
- Reinforcement Learning with Unsupervised Auxiliary Tasks
- The Consciousness Prior
- BabyAI: A Platform to Study the Sample Efficiency of Grounded Language Learning
- Investigating Simple Object Representations in Model-Free Deep Reinforcement Learning
- Learning Object-Centric Video Models by Contrasting Sets