Learning Belief Representations for Imitation Learning in POMDPs

Abstract

We consider the problem of imitation learning from expert demonstrations in partially observable Markov decision processes (POMDPs). Belief representations, which characterize the distribution over the latent states in a POMDP, have been modeled using recurrent neural networks and probabilistic latent variable models, and shown to be effective for reinforcement learning in POMDPs. In this work, we investigate the belief representation learning problem for generative adversarial imitation learning in POMDPs. Instead of training the belief module and the policy separately as suggested in prior work, we learn the belief module jointly with the policy, using a task-aware imitation loss to ensure that the representation is more aligned with the policy's objective. To improve robustness of representation, we introduce several informative belief regularization techniques, including multi-step prediction of dynamics and action-sequences. Evaluated on various partially observable continuous-control locomotion tasks, our belief-module imitation learning approach (BMIL) substantially outperforms several baselines, including the original GAIL algorithm and the task-agnostic belief learning algorithm. Extensive ablation analysis indicates the effectiveness of task-aware belief learning and belief regularization.

Conference on Uncertainty in Artificial Intelligence (UAI 2019)

References in corpus (10)

• Trust Region Policy Optimization
• High-Dimensional Continuous Control Using Generalized Advantage Estimation
• Variational Inference with Normalizing Flows
• Value-Function Approximations for Partially Observable Markov Decision Processes
• Online Planning Algorithms for POMDPs
• Recurrent World Models Facilitate Policy Evolution
• Gradient Estimation Using Stochastic Computation Graphs
• Reinforcement and Imitation Learning via Interactive No-Regret Learning
• Neural Predictive Belief Representations
• Recurrent Predictive State Policy Networks

Cited by in corpus (2)

• Semi-Supervised Imitation Learning of Team Policies from Suboptimal Demonstrations
• Approximate Model-Based Shielding for Safe Reinforcement Learning