تقرير
MVMoE: Multi-Task Vehicle Routing Solver with Mixture-of-Experts
| العنوان: | MVMoE: Multi-Task Vehicle Routing Solver with Mixture-of-Experts |
|---|---|
| المؤلفون: | Zhou, Jianan, Cao, Zhiguang, Wu, Yaoxin, Song, Wen, Ma, Yining, Zhang, Jie, Xu, Chi |
| سنة النشر: | 2024 |
| المجموعة: | Computer Science |
| مصطلحات موضوعية: | Computer Science - Artificial Intelligence, Computer Science - Machine Learning |
| الوصف: | Learning to solve vehicle routing problems (VRPs) has garnered much attention. However, most neural solvers are only structured and trained independently on a specific problem, making them less generic and practical. In this paper, we aim to develop a unified neural solver that can cope with a range of VRP variants simultaneously. Specifically, we propose a multi-task vehicle routing solver with mixture-of-experts (MVMoE), which greatly enhances the model capacity without a proportional increase in computation. We further develop a hierarchical gating mechanism for the MVMoE, delivering a good trade-off between empirical performance and computational complexity. Experimentally, our method significantly promotes zero-shot generalization performance on 10 unseen VRP variants, and showcases decent results on the few-shot setting and real-world benchmark instances. We further conduct extensive studies on the effect of MoE configurations in solving VRPs, and observe the superiority of hierarchical gating when facing out-of-distribution data. The source code is available at: https://github.com/RoyalSkye/Routing-MVMoETest. Comment: Accepted at ICML 2024 |
| نوع الوثيقة: | Working Paper |
| الوصول الحر: | http://arxiv.org/abs/2405.01029Test |
| رقم الانضمام: | edsarx.2405.01029 |
| قاعدة البيانات: | arXiv |
| الوصف غير متاح. |