EGAM: Extended Graph Attention Model for Solving Routing Problems

TL;DR

EGAM extends the graph attention model by incorporating edge features and multi-head attention, significantly improving neural combinatorial optimization for routing problems, especially in complex scenarios.

Contribution

The paper introduces EGAM, a novel graph attention model that updates both node and edge embeddings, enhancing performance over existing models in routing tasks.

Findings

EGAM matches or outperforms existing methods on various routing problems.

EGAM excels in highly constrained and complex graph structures.

The model demonstrates strong generalization across different problem types.

Abstract

Neural combinatorial optimization (NCO) solvers, implemented with graph neural networks (GNNs), have introduced new approaches for solving routing problems. Trained with reinforcement learning (RL), the state-of-the-art graph attention model (GAM) achieves near-optimal solutions without requiring expert knowledge or labeled data. In this work, we generalize the existing graph attention mechanism and propose the extended graph attention model (EGAM). Our model utilizes multi-head dot-product attention to update both node and edge embeddings, addressing the limitations of the conventional GAM, which considers only node features. We employ an autoregressive encoder-decoder architecture and train it with policy gradient algorithms that incorporate a specially designed baseline. Experiments show that EGAM matches or outperforms existing methods across various routing problems. Notably, the proposed model demonstrates exceptional performance on highly constrained problems, highlighting its efficiency in handling complex graph structures.