Uncertainty Representations in State-Space Layers for Deep Reinforcement Learning under Partial Observability
This addresses the challenge of partial observability in reinforcement learning for agents, though it is incremental as it builds on existing probabilistic world models.
The paper tackled the problem of decision-making under partial observability in reinforcement learning by proposing a Kalman filter layer that incorporates uncertainty into state representations, resulting in improved performance over other stateful models in tasks where uncertainty reasoning is critical.
Optimal decision-making under partial observability requires reasoning about the uncertainty of the environment's hidden state. However, most reinforcement learning architectures handle partial observability with sequence models that have no internal mechanism to incorporate uncertainty in their hidden state representation, such as recurrent neural networks, deterministic state-space models and transformers. Inspired by advances in probabilistic world models for reinforcement learning, we propose a standalone Kalman filter layer that performs closed-form Gaussian inference in linear state-space models and train it end-to-end within a model-free architecture to maximize returns. Similar to efficient linear recurrent layers, the Kalman filter layer processes sequential data using a parallel scan, which scales logarithmically with the sequence length. By design, Kalman filter layers are a drop-in replacement for other recurrent layers in standard model-free architectures, but importantly they include an explicit mechanism for probabilistic filtering of the latent state representation. Experiments in a wide variety of tasks with partial observability show that Kalman filter layers excel in problems where uncertainty reasoning is key for decision-making, outperforming other stateful models.