DeepAI AI Chat
Log In Sign Up

Invariant Causal Prediction for Block MDPs

by   Amy Zhang, et al.

Generalization across environments is critical to the successful application of reinforcement learning algorithms to real-world challenges. In this paper, we consider the problem of learning abstractions that generalize in block MDPs, families of environments with a shared latent state space and dynamics structure over that latent space, but varying observations. We leverage tools from causal inference to propose a method of invariant prediction to learn model-irrelevance state abstractions (MISA) that generalize to novel observations in the multi-environment setting. We prove that for certain classes of environments, this approach outputs with high probability a state abstraction corresponding to the causal feature set with respect to the return. We further provide more general bounds on model error and generalization error in the multi-environment setting, in the process showing a connection between causal variable selection and the state abstraction framework for MDPs. We give empirical evidence that our methods work in both linear and nonlinear settings, attaining improved generalization over single- and multi-task baselines.


page 1

page 2

page 3

page 4


Model-Invariant State Abstractions for Model-Based Reinforcement Learning

Accuracy and generalization of dynamics models is key to the success of ...

Multi-Task Reinforcement Learning as a Hidden-Parameter Block MDP

Multi-task reinforcement learning is a rich paradigm where information f...

Multi-Step Prediction in Linearized Latent State Spaces for Representation Learning

In this paper, we derive a novel method as a generalization over LCEs su...

Randomized Value Functions via Posterior State-Abstraction Sampling

State abstraction has been an essential tool for dramatically improving ...

Learning Invariant Representations for Reinforcement Learning without Reconstruction

We study how representation learning can accelerate reinforcement learni...

Lower Bounds on the Error Probability for Invariant Causal Prediction

It is common practice to collect observations of feature and response pa...

Jointly Learning Consistent Causal Abstractions Over Multiple Interventional Distributions

An abstraction can be used to relate two structural causal models repres...