Learning Parameterized Skills from Demonstrations
This work addresses the challenge of skill learning in robotics and AI for better multitask generalization, though it appears incremental as it builds on existing skill learning methods.
The paper tackles the problem of discovering parameterized skills from expert demonstrations by introducing DEPS, an end-to-end algorithm that learns skill policies and a meta-policy jointly, resulting in improved generalization to unseen tasks and outperforming baselines on LIBERO and MetaWorld benchmarks.
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.