paper-with-me

Papers

Predicate Invention for Bilevel Planning

2022-03-17 · Tom Silver, Rohan Chitnis, Nishanth Kumar, Willie McClinton, Tomas Lozano-Perez, Leslie Pack Kaelbling, Joshua Tenenbaum

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

📄 PDF Abstract BibTeX arXiv:2203.09634

Code (1)

Learning-and-Intelligent-Systems/predicators 공식 구현

Similar Papers 제목 키워드 기반

SkillWrapper: Generative Predicate Invention for Task-level Robot Planning

2025-11-22 · Ziyi Yang, Benned Hedegaard, Ahmed Jaafar, Yichen Wei 외 arxiv

Generalizing from individual skill executions to long-horizon tasks is a core challenge in building autonomous robots. A promising direction is learning high-level, symbolic representations of low-level robot skills, ena…

Representation Learning

Unifying Deep Predicate Invention with Pre-trained Foundation Models

2025-12-19 · Qianwei Wang, Bowen Li, Zhanpeng Luo, Yifan Xu 외 arxiv

Long-horizon robotic tasks are hard due to continuous state-action spaces and sparse feedback. Symbolic world models help by decomposing tasks into discrete predicates that capture object properties and relations. Existi…

Theory reconstruction: a representation learning view on predicate invention

2016-06-28 · Sebastijan Dumancic, Wannes Meert, Hendrik Blockeel

With this positional paper we present a representation learning view on predicate invention. The intention of this proposal is to bridge the relational and deep learning communities on the problem of predicate invention.…

Representation Learning

Embodied Active Learning of Relational State Abstractions for Bilevel Planning

2023-03-08 · Amber Li, Tom Silver

State abstraction is an effective technique for planning in robotics environments with continuous states and actions, long task horizons, and sparse feedback. In object-oriented environments, predicates are a particularl…

Active LearningInformativeness

VisualPredicator: Learning Abstract World Models with Neuro-Symbolic Predicates for Robot Planning

2024-10-30 · Yichao Liang, Nishanth Kumar, Hao Tang, Adrian Weller 외

Broadly intelligent agents should form task-specific abstractions that selectively expose the essential elements of a task, while abstracting away the complexity of the raw sensorimotor space. In this work, we present Ne…

Hierarchical Reinforcement LearningLanguage ModelingLanguage ModellingOut-of-Distribution Generalization