Neural Predictive Belief Representations

TL;DR

This paper explores neural methods for learning belief state representations in partially observable environments, demonstrating that contrastive predictive coding can encode both state and uncertainty, aiding decision making.

Contribution

It shows that neural architectures using CPC can effectively learn belief representations, including uncertainty, in complex environments, advancing unsupervised learning for partially observable domains.

Findings

All methods learned belief representations encoding environment state.

CPC with multi-step prediction and action-conditioning improves accuracy.

Representations include both state information and uncertainty.

Abstract

Unsupervised representation learning has succeeded with excellent results in many applications. It is an especially powerful tool to learn a good representation of environments with partial or noisy observations. In partially observable domains it is important for the representation to encode a belief state, a sufficient statistic of the observations seen so far. In this paper, we investigate whether it is possible to learn such a belief representation using modern neural architectures. Specifically, we focus on one-step frame prediction and two variants of contrastive predictive coding (CPC) as the objective functions to learn the representations. To evaluate these learned representations, we test how well they can predict various pieces of information about the underlying state of the environment, e.g., position of the agent in a 3D maze. We show that all three methods are able to learn belief representations of the environment, they encode not only the state information, but also its uncertainty, a crucial aspect of belief states. We also find that for CPC multi-step predictions and action-conditioning are critical for accurate belief representations in visually complex environments. The ability of neural representations to capture the belief information has the potential to spur new advances for learning and planning in partially observable domains, where leveraging uncertainty is essential for optimal decision making.