Differentiable Learning of Lifted Action Schemas for Classical Planning

TL;DR

This paper introduces a neural network architecture for learning lifted action schemas from fully observed states with unobserved action arguments, advancing the ability to learn planning domains from sequences.

Contribution

The work presents a novel differentiable model that learns action schemas and identifies action arguments from state change traces, facilitating neuro-symbolic planning domain learning.

Findings

Successfully recovers ground-truth structure of action schemas in various planning domains.

Demonstrates robustness to observation noise.

Effectively learns slot-based dynamics models.

Abstract

Classical planners can effectively solve very large deterministic MDPs represented in STRIPS or PDDL where states are sets of atoms over objects and relations, and lifted action schemas add or delete these atoms. This compact representation yields strong search heuristics and provides an ideal setting for structural generalization, since lifted relations and action schemas give rise to infinitely many domain instances. A central challenge is to learn these relations and action schemas from data, and recent approaches have addressed this problem using different types of observations. In this work, we develop a novel neural network architecture for learning action schemas from traces where states are fully observed but action arguments are unobserved. The problem is a simplification but an important step towards learning planning domains from sequences of images and action labels, and we aim to solve this simplification in a nearly perfect manner. The challenge lies in learning the action schemas while simultaneously identifying the action arguments from observed state changes. Our approach yields a robust differentiable component that can then be integrated into larger neuro-symbolic models. We evaluate the architecture on various planning domains, where the learned lifted action schemas must recover the ground-truth structure. Additionally, we report experiments on robustness to observation noise and on a variation related to slot-based dynamics models.