Differentiable Learning of Lifted Action Schemas for Classical Planning
Summary
This paper introduces a neural network architecture that learns lifted action schemas from fully observed state traces with unobserved action arguments, aiming to enable robust learning of planning domains for neuro-symbolic models.
View Cached Full Text
Cached at: 05/14/26, 06:15 AM
# Differentiable Learning of Lifted Action Schemas for Classical Planning Source: [https://arxiv.org/abs/2605.13282](https://arxiv.org/abs/2605.13282) [View PDF](https://arxiv.org/pdf/2605.13282) > 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\. ## Submission history From: Jonas Reiter \[[view email](https://arxiv.org/show-email/ce83a3d6/2605.13282)\] **\[v1\]**Wed, 13 May 2026 09:59:49 UTC \(374 KB\)
Similar Articles
Learning Bilevel Policies over Symbolic World Models for Long-Horizon Planning
Proposes BISON, a system combining learned low-level neural policies with high-level symbolic planning for long-horizon embodied tasks, showing strong generalization and efficiency.
Neuro-Inspired Inverse Learning for Planning and Control
This paper introduces a neuro-inspired framework called Inverter that uses Inverse Learning (IL) for fast and efficient planning and control, achieving significant improvements on D4RL benchmarks and quantum gate synthesis with orders of magnitude less inference computation.
Learning and Reusing Policy Decompositions for Hierarchical Generalized Planning with LLM Agents
This paper introduces HCL-GP, a dynamic policy-learning framework that integrates generalized planning and hierarchical task decomposition to enable LLM-based agents to learn and reuse executable policy components, significantly improving performance on the AppWorld benchmark.
Shaping Schema via Language Representation as the Next Frontier for LLM Intelligence Expanding
This paper argues that designing advanced language representations to shape cognitive schemas is a key frontier for expanding LLM intelligence without scaling parameters. It provides formalizations and empirical evidence showing that different linguistic structures significantly impact model performance and internal feature activations.
Agentic Transformers Provably Learn to Search via Reinforcement Learning
This paper theoretically studies how transformer-based policies acquire search capabilities from reinforcement learning training dynamics in a stochastic tree environment. It shows that a two-head transformer can implement depth-first search and that this mechanism emerges naturally from sparse reward signals under a depth-wise curriculum.