RouteFinder: Towards Foundation Models for Vehicle Routing Problems
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915494462750720 |
|---|---|
| author | Berto, Federico Hua, Chuanbo Zepeda, Nayeli Gast Hottung, André Wouda, Niels Lan, Leon Park, Junyoung Tierney, Kevin Park, Jinkyoo |
| author_facet | Berto, Federico Hua, Chuanbo Zepeda, Nayeli Gast Hottung, André Wouda, Niels Lan, Leon Park, Junyoung Tierney, Kevin Park, Jinkyoo |
| contents | This paper introduces RouteFinder, a comprehensive foundation model framework to tackle different Vehicle Routing Problem (VRP) variants. Our core idea is that a foundation model for VRPs should be able to represent variants by treating each as a subset of a generalized problem equipped with different attributes. We propose a unified VRP environment capable of efficiently handling any combination of these attributes. The RouteFinder model leverages a modern transformer-based encoder and global attribute embeddings to improve task representation. Additionally, we introduce two reinforcement learning techniques to enhance multi-task performance: mixed batch training, which enables training on different variants at once, and multi-variant reward normalization to balance different reward scales. Finally, we propose efficient adapter layers that enable fine-tuning for new variants with unseen attributes. Extensive experiments on 48 VRP variants show RouteFinder outperforms recent state-of-the-art learning methods. Our code is publicly available at https://github.com/ai4co/routefinder. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_15007 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | RouteFinder: Towards Foundation Models for Vehicle Routing Problems Berto, Federico Hua, Chuanbo Zepeda, Nayeli Gast Hottung, André Wouda, Niels Lan, Leon Park, Junyoung Tierney, Kevin Park, Jinkyoo Artificial Intelligence This paper introduces RouteFinder, a comprehensive foundation model framework to tackle different Vehicle Routing Problem (VRP) variants. Our core idea is that a foundation model for VRPs should be able to represent variants by treating each as a subset of a generalized problem equipped with different attributes. We propose a unified VRP environment capable of efficiently handling any combination of these attributes. The RouteFinder model leverages a modern transformer-based encoder and global attribute embeddings to improve task representation. Additionally, we introduce two reinforcement learning techniques to enhance multi-task performance: mixed batch training, which enables training on different variants at once, and multi-variant reward normalization to balance different reward scales. Finally, we propose efficient adapter layers that enable fine-tuning for new variants with unseen attributes. Extensive experiments on 48 VRP variants show RouteFinder outperforms recent state-of-the-art learning methods. Our code is publicly available at https://github.com/ai4co/routefinder. |
| title | RouteFinder: Towards Foundation Models for Vehicle Routing Problems |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2406.15007 |