Learning Parameterized Skills from Demonstrations

TL;DR

This paper introduces DEPS, an end-to-end algorithm that learns interpretable, parameterized skills from expert demonstrations, improving generalization to new tasks and outperforming existing baselines on benchmark datasets.

Contribution

DEPS jointly learns parameterized skill policies and a meta-policy using temporal variational inference, addressing latent variable degeneracy and enhancing skill interpretability.

Findings

Outperforms baseline methods on LIBERO and MetaWorld benchmarks.

Learns semantically meaningful, temporally extended skills.

Discovers interpretable skills with continuous parameters, like grasp locations.

Abstract

We present DEPS, an end-to-end algorithm for discovering parameterized skills from expert demonstrations. Our method learns parameterized skill policies jointly with a meta-policy that selects the appropriate discrete skill and continuous parameters at each timestep. Using a combination of temporal variational inference and information-theoretic regularization methods, we address the challenge of degeneracy common in latent variable models, ensuring that the learned skills are temporally extended, semantically meaningful, and adaptable. We empirically show that learning parameterized skills from multitask expert demonstrations significantly improves generalization to unseen tasks. Our method outperforms multitask as well as skill learning baselines on both LIBERO and MetaWorld benchmarks. We also demonstrate that DEPS discovers interpretable parameterized skills, such as an object grasping skill whose continuous arguments define the grasp location.