pith. sign in

arxiv: 2007.05838 · v1 · pith:I4LH3MNBnew · submitted 2020-07-11 · 💻 cs.LG · cs.AI· stat.ML

Control as Hybrid Inference

classification 💻 cs.LG cs.AIstat.ML
keywords inferencemodel-basedmodel-freealgorithmcontrolamortisedframeworkhybrid
0
0 comments X
read the original abstract

The field of reinforcement learning can be split into model-based and model-free methods. Here, we unify these approaches by casting model-free policy optimisation as amortised variational inference, and model-based planning as iterative variational inference, within a `control as hybrid inference' (CHI) framework. We present an implementation of CHI which naturally mediates the balance between iterative and amortised inference. Using a didactic experiment, we demonstrate that the proposed algorithm operates in a model-based manner at the onset of learning, before converging to a model-free algorithm once sufficient data have been collected. We verify the scalability of our algorithm on a continuous control benchmark, demonstrating that it outperforms strong model-free and model-based baselines. CHI thus provides a principled framework for harnessing the sample efficiency of model-based planning while retaining the asymptotic performance of model-free policy optimisation.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.