Predicate Invention for Bilevel Planning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2022
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912396171280384 |
|---|---|
| author | Silver, Tom Chitnis, Rohan Kumar, Nishanth McClinton, Willie Lozano-Perez, Tomas Kaelbling, Leslie Pack Tenenbaum, Joshua |
| author_facet | Silver, Tom Chitnis, Rohan Kumar, Nishanth McClinton, Willie Lozano-Perez, Tomas Kaelbling, Leslie Pack Tenenbaum, Joshua |
| contents | Efficient planning in continuous state and action spaces is fundamentally hard, even when the transition model is deterministic and known. One way to alleviate this challenge is to perform bilevel planning with abstractions, where a high-level search for abstract plans is used to guide planning in the original transition space. Previous work has shown that when state abstractions in the form of symbolic predicates are hand-designed, operators and samplers for bilevel planning can be learned from demonstrations. In this work, we propose an algorithm for learning predicates from demonstrations, eliminating the need for manually specified state abstractions. Our key idea is to learn predicates by optimizing a surrogate objective that is tractable but faithful to our real efficient-planning objective. We use this surrogate objective in a hill-climbing search over predicate sets drawn from a grammar. Experimentally, we show across four robotic planning environments that our learned abstractions are able to quickly solve held-out tasks, outperforming six baselines. Code: https://tinyurl.com/predicators-release |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2203_09634 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Predicate Invention for Bilevel Planning Silver, Tom Chitnis, Rohan Kumar, Nishanth McClinton, Willie Lozano-Perez, Tomas Kaelbling, Leslie Pack Tenenbaum, Joshua Artificial Intelligence Machine Learning Robotics Efficient planning in continuous state and action spaces is fundamentally hard, even when the transition model is deterministic and known. One way to alleviate this challenge is to perform bilevel planning with abstractions, where a high-level search for abstract plans is used to guide planning in the original transition space. Previous work has shown that when state abstractions in the form of symbolic predicates are hand-designed, operators and samplers for bilevel planning can be learned from demonstrations. In this work, we propose an algorithm for learning predicates from demonstrations, eliminating the need for manually specified state abstractions. Our key idea is to learn predicates by optimizing a surrogate objective that is tractable but faithful to our real efficient-planning objective. We use this surrogate objective in a hill-climbing search over predicate sets drawn from a grammar. Experimentally, we show across four robotic planning environments that our learned abstractions are able to quickly solve held-out tasks, outperforming six baselines. Code: https://tinyurl.com/predicators-release |
| title | Predicate Invention for Bilevel Planning |
| topic | Artificial Intelligence Machine Learning Robotics |
| url | https://arxiv.org/abs/2203.09634 |