We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Computer Science > Machine Learning

Title: LISA: Learning Interpretable Skill Abstractions from Language

Abstract: Learning policies that effectively utilize language instructions in complex, multi-task environments is an important problem in sequential decision-making. While it is possible to condition on the entire language instruction directly, such an approach could suffer from generalization issues. In our work, we propose \emph{Learning Interpretable Skill Abstractions (LISA)}, a hierarchical imitation learning framework that can learn diverse, interpretable primitive behaviors or skills from language-conditioned demonstrations to better generalize to unseen instructions. LISA uses vector quantization to learn discrete skill codes that are highly correlated with language instructions and the behavior of the learned policy. In navigation and robotic manipulation environments, LISA outperforms a strong non-hierarchical Decision Transformer baseline in the low data regime and is able to compose learned skills to solve tasks containing unseen long-range instructions. Our method demonstrates a more natural way to condition on language in sequential decision-making problems and achieve interpretable and controllable behavior with the learned skills.
Comments: NeurIPS 2022. Website: this https URL
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Robotics (cs.RO)
Cite as: arXiv:2203.00054 [cs.LG]
  (or arXiv:2203.00054v3 [cs.LG] for this version)

Submission history

From: Divyansh Garg [view email]
[v1] Mon, 28 Feb 2022 19:43:24 GMT (10419kb,D)
[v2] Tue, 15 Nov 2022 19:30:37 GMT (10404kb,D)
[v3] Tue, 6 Dec 2022 05:20:36 GMT (10404kb,D)

Link back to: arXiv, form interface, contact.