ERL-MPP: Evolutionary Reinforcement Learning with Multi-head Puzzle Perception for Solving Large-scale Jigsaw Puzzles of Eroded Gaps

2025-04-11

AAAI Conference on Artificial Intelligence

AI Summary

To address local misalignment and global structural instability caused by corrosive inter-piece gaps in large-scale jigsaw puzzle reconstruction, this paper proposes a joint framework comprising a Multi-Head Puzzle-Aware Network (MPPN) and Evolutionary Reinforcement Learning (EvoRL). MPPN employs a shared encoder with multiple puzzle-heads to extract gap-robust features, while integrating a discriminative head and an Actor-Critic architecture to model local assembly states. EvoRL introduces a history-aware evolutionary policy evaluator to efficiently search the ultra-large swap-action space. Our method achieves state-of-the-art performance on both JPLEG-5 (large-gap) and MIT (large-scale) benchmarks, significantly outperforming prior approaches in reconstruction accuracy and gap robustness. Notably, it is the first end-to-end learnable framework capable of reconstructing large-scale puzzles under severe corrosive gap conditions.

Technology Category

• Machine Learning: Evolutionary Learning
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work

Abstract

Solving jigsaw puzzles has been extensively studied. While most existing models focus on solving either small-scale puzzles or puzzles with no gap between fragments, solving large-scale puzzles with gaps presents distinctive challenges in both image understanding and combinatorial optimization. To tackle these challenges, we propose a framework of Evolutionary Reinforcement Learning with Multi-head Puzzle Perception (ERL-MPP) to derive a better set of swapping actions for solving the puzzles. Specifically, to tackle the challenges of perceiving the puzzle with gaps, a Multi-head Puzzle Perception Network (MPPN) with a shared encoder is designed, where multiple puzzlet heads comprehensively perceive the local assembly status, and a discriminator head provides a global assessment of the puzzle. To explore the large swapping action space efficiently, an Evolutionary Reinforcement Learning (EvoRL) agent is designed, where an actor recommends a set of suitable swapping actions from a large action space based on the perceived puzzle status, a critic updates the actor using the estimated rewards and the puzzle status, and an evaluator coupled with evolutionary strategies evolves the actions aligning with the historical assembly experience. The proposed ERL-MPP is comprehensively evaluated on the JPLEG-5 dataset with large gaps and the MIT dataset with large-scale puzzles. It significantly outperforms all state-of-the-art models on both datasets.

Problem

• Solving large-scale jigsaw puzzles with eroded gaps
• Challenges in image understanding and combinatorial optimization
• Efficiently exploring large swapping action space

Innovation

• Multi-head Puzzle Perception Network for gap perception
• Evolutionary Reinforcement Learning for action exploration
• Shared encoder with discriminator for global assessment