Advancing Learnable Multi-Agent Pathfinding Solvers with Active Fine-Tuning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Andreychuk, Anton, Yakovlev, Konstantin, Panov, Aleksandr, Skrynnik, Alexey
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912457747857408
author Andreychuk, Anton
Yakovlev, Konstantin
Panov, Aleksandr
Skrynnik, Alexey
author_facet Andreychuk, Anton
Yakovlev, Konstantin
Panov, Aleksandr
Skrynnik, Alexey
contents Multi-agent pathfinding (MAPF) is a common abstraction of multi-robot trajectory planning problems, where multiple homogeneous robots simultaneously move in the shared environment. While solving MAPF optimally has been proven to be NP-hard, scalable, and efficient, solvers are vital for real-world applications like logistics, search-and-rescue, etc. To this end, decentralized suboptimal MAPF solvers that leverage machine learning have come on stage. Building on the success of the recently introduced MAPF-GPT, a pure imitation learning solver, we introduce MAPF-GPT-DDG. This novel approach effectively fine-tunes the pre-trained MAPF model using centralized expert data. Leveraging a novel delta-data generation mechanism, MAPF-GPT-DDG accelerates training while significantly improving performance at test time. Our experiments demonstrate that MAPF-GPT-DDG surpasses all existing learning-based MAPF solvers, including the original MAPF-GPT, regarding solution quality across many testing scenarios. Remarkably, it can work with MAPF instances involving up to 1 million agents in a single environment, setting a new milestone for scalability in MAPF domains.
format Preprint
id arxiv_https___arxiv_org_abs_2506_23793
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Advancing Learnable Multi-Agent Pathfinding Solvers with Active Fine-Tuning
Andreychuk, Anton
Yakovlev, Konstantin
Panov, Aleksandr
Skrynnik, Alexey
Artificial Intelligence
Machine Learning
Multiagent Systems
Multi-agent pathfinding (MAPF) is a common abstraction of multi-robot trajectory planning problems, where multiple homogeneous robots simultaneously move in the shared environment. While solving MAPF optimally has been proven to be NP-hard, scalable, and efficient, solvers are vital for real-world applications like logistics, search-and-rescue, etc. To this end, decentralized suboptimal MAPF solvers that leverage machine learning have come on stage. Building on the success of the recently introduced MAPF-GPT, a pure imitation learning solver, we introduce MAPF-GPT-DDG. This novel approach effectively fine-tunes the pre-trained MAPF model using centralized expert data. Leveraging a novel delta-data generation mechanism, MAPF-GPT-DDG accelerates training while significantly improving performance at test time. Our experiments demonstrate that MAPF-GPT-DDG surpasses all existing learning-based MAPF solvers, including the original MAPF-GPT, regarding solution quality across many testing scenarios. Remarkably, it can work with MAPF instances involving up to 1 million agents in a single environment, setting a new milestone for scalability in MAPF domains.
title Advancing Learnable Multi-Agent Pathfinding Solvers with Active Fine-Tuning
topic Artificial Intelligence
Machine Learning
Multiagent Systems
url https://arxiv.org/abs/2506.23793