MTL-KD: Multi-Task Learning Via Knowledge Distillation for Generalizable Neural Vehicle Routing Solver

TL;DR

This paper introduces MTL-KD, a knowledge distillation-based multi-task learning approach for neural vehicle routing, enabling training of large, generalizable models that perform well on diverse and large-scale VRP variants.

Contribution

The paper presents a novel MTL-KD method that transfers knowledge from single-task RL models to a heavy decoder, improving generalization across multiple VRP variants.

Findings

Achieves superior performance on 6 seen and 10 unseen VRP variants.

Demonstrates strong generalization on large-scale problems with up to 1000 nodes.

Introduces R3C inference strategy to enhance multi-task model performance.

Abstract

Multi-Task Learning (MTL) in Neural Combinatorial Optimization (NCO) is a promising approach to train a unified model capable of solving multiple Vehicle Routing Problem (VRP) variants. However, existing Reinforcement Learning (RL)-based multi-task methods can only train light decoder models on small-scale problems, exhibiting limited generalization ability when solving large-scale problems. To overcome this limitation, this work introduces a novel multi-task learning method driven by knowledge distillation (MTL-KD), which enables the efficient training of heavy decoder models with strong generalization ability. The proposed MTL-KD method transfers policy knowledge from multiple distinct RL-based single-task models to a single heavy decoder model, facilitating label-free training and effectively improving the model's generalization ability across diverse tasks. In addition, we introduce a flexible inference strategy termed Random Reordering Re-Construction (R3C), which is specifically adapted for diverse VRP tasks and further boosts the performance of the multi-task model. Experimental results on 6 seen and 10 unseen VRP variants with up to 1000 nodes indicate that our proposed method consistently achieves superior performance on both uniform and real-world benchmarks, demonstrating robust generalization abilities.