Publication Type
Conference Proceeding Article
Version
publishedVersion
Publication Date
6-2022
Abstract
Recent deep models for solving routing problems always assume a single distribution of nodes for training, which severely impairs their cross-distribution generalization ability. In this paper, we exploit group distributionally robust optimization (group DRO) to tackle this issue, where we jointly optimize the weights for different groups of distributions and the parameters for the deep model in an interleaved manner during training. We also design a module based on convolutional neural network, which allows the deep model to learn more informative latent pattern among the nodes. We evaluate the proposed approach on two types of wellknown deep models including GCN and POMO. The experimental results on the randomly synthesized instances and the ones from two benchmark dataset (i.e., TSPLib and CVRPLib) demonstrate that our approach could significantly improve the cross-distribution generalization performance over the original models.
Keywords
Benchmark datasets, Convolutional neural network, Generalization ability, Generalization performance, Learn+, Module-based, Original model, Robust optimization, Routing problems, Synthesised
Discipline
Software Engineering | Theory and Algorithms
Publication
Proceedings of the 36th AAAI Conference on Artificial Intelligence, AAAI 2022, Virtual Online, Feb 22- Mar 1
Volume
36
First Page
9786
Last Page
9794
ISBN
9781577358763
Publisher
Association for the Advancement of Artificial Intelligence
City or Country
Virtual, Online
Citation
YUAN, Jiang; WU, Yaoxin; and CAO, Zhiguang.
Learning to solve routing problems via distributionally robust optimization. (2022). Proceedings of the 36th AAAI Conference on Artificial Intelligence, AAAI 2022, Virtual Online, Feb 22- Mar 1. 36, 9786-9794.
Available at: https://ink.library.smu.edu.sg/sis_research/8162
Copyright Owner and License
Authors
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.