Learning Robot Skills with Temporal Variational Inference
This addresses the problem of enabling robots to autonomously learn hierarchical skills from demonstrations, which is incremental as it builds on existing variational inference methods.
The paper tackles unsupervised discovery of robotic options from demonstrations by jointly learning low-level control policies and higher-level policies using continuous latent variables, achieving successful learning across three robotic datasets.
In this paper, we address the discovery of robotic options from demonstrations in an unsupervised manner. Specifically, we present a framework to jointly learn low-level control policies and higher-level policies of how to use them from demonstrations of a robot performing various tasks. By representing options as continuous latent variables, we frame the problem of learning these options as latent variable inference. We then present a temporal formulation of variational inference based on a temporal factorization of trajectory likelihoods,that allows us to infer options in an unsupervised manner. We demonstrate the ability of our framework to learn such options across three robotic demonstration datasets.