planning problem
- North America > United States > Arizona > Maricopa County > Tempe (0.04)
- North America > United States > Colorado > Larimer County > Fort Collins (0.04)
- Europe > Czechia > Prague (0.04)
- North America > United States > Arizona (0.04)
- North America > United States > Colorado (0.04)
- Europe > United Kingdom > England > Cambridgeshire > Cambridge (0.04)
- Europe > Czechia > Prague (0.04)
- Research Report > New Finding (1.00)
- Research Report > Experimental Study (0.68)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Planning & Scheduling (0.97)
- Information Technology > Artificial Intelligence > Natural Language (0.68)
- Information Technology > Artificial Intelligence > Machine Learning > Learning Graphical Models > Undirected Networks > Markov Models (0.49)
- North America > United States > Minnesota > Hennepin County > Minneapolis (0.14)
- North America > Canada > Quebec > Montreal (0.04)
- North America > Canada > Ontario > Kingston (0.04)
- (2 more...)
- Overview (0.67)
- Research Report > New Finding (0.67)
- North America > United States > Arizona (0.05)
- Asia > Middle East > Jordan (0.04)
- North America > United States > Colorado (0.04)
- (2 more...)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Planning & Scheduling (1.00)
- Information Technology > Artificial Intelligence > Natural Language > Large Language Model (1.00)
- Information Technology > Artificial Intelligence > Machine Learning (1.00)
- Information Technology > Artificial Intelligence > Cognitive Science > Problem Solving (0.94)
What Planning Problems Can A Relational Neural Network Solve?
Goal-conditioned policies are generally understood to be feed-forward circuits, in the form of neural networks that map from the current state and the goal specification to the next action to take. However, under what circumstances such a policy can be learned and how efficient the policy will be are not well understood. In this paper, we present a circuit complexity analysis for relational neural networks (such as graph neural networks and transformers) representing policies for planning problems, by drawing connections with serialized goal regression search (S-GRS). We show that there are three general classes of planning problems, in terms of the growth of circuit width and depth as a function of the number of objects and planning horizon, providing constructive proofs. We also illustrate the utility of this analysis for designing neural networks for policy learning.
Leveraging Environment Interaction for Automated PDDL Translation and Planning with Large Language Models
Large Language Models (LLMs) have shown remarkable performance in various natural language tasks, but they often struggle with planning problems that require structured reasoning. To address this limitation, the conversion of planning problems into the Planning Domain Definition Language (PDDL) has been proposed as a potential solution, enabling the use of automated planners. However, generating accurate PDDL files typically demands human inputs or correction, which can be time-consuming and costly. In this paper, we propose a novel approach that leverages LLMs and environment feedback to automatically generate PDDL domain and problem description files without the need for human intervention. Our method introduces an iterative refinement process that generates multiple problem PDDL candidates and progressively refines the domain PDDL based on feedback obtained from interacting with the environment. To guide the refinement process, we develop an Exploration Walk (EW) metric, which provides rich feedback signals for LLMs to update the PDDL file. We evaluate our approach on $10$ PDDL environments. We achieve an average task solve rate of 66\% compared to a 29\% solve rate by GPT-4's intrinsic planning with chain-of-thought prompting. Our work enables the automated modeling of planning environments using LLMs and environment feedback, eliminating the need for human intervention in the PDDL translation process and paving the way for more reliable LLM agents in challenging problems.