Predicate Invention for Bilevel Planning
Tom Silver\equalcontrib1, Rohan Chitnis\equalcontrib2, Nishanth Kumar1, Willie McClinton1, Tomás Lozano-Pérez1, Leslie Pack Kaelbling1, Joshua B. Tenenbaum1
Abstract
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.
中文速览
机器人规划任务里，人类工程师需要手工设计符号谓词（symbolic predicates）来描述状态抽象，这既费力又限制了系统的通用性。本文提出一种算法，让机器人直接从演示数据中自动学习这些谓词，核心思路是设计一个既忠实于"双层规划效率"这一真实目标、又计算上可处理的代理目标函数（surrogate objective），再借助语法制导的爬山搜索从候选谓词空间中筛选出最优谓词集合。在四个机器人规划环境中的实验表明，用50–200条演示学到的谓词、算子和采样器，能高效解决训练中未见过的新任务，明显优于六种基线方法。这项工作填补了任务与运动规划中状态抽象须靠人工设计的关键缺口，为端到端自动构建层次规划所需的符号表示迈出了重要一步。
原文 arXiv:2203.09634;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2203.09634v3