Step-wise deep learning models for solving routing problems
Routing problems are very important in intelligent transportation systems. Recently, a number of deep learning-based methods are proposed to automatically learn construction heuristics for solving routing problems. However, these methods do not completely follow Bellman's Principle of Optimalit...
Saved in:
Main Authors: | , , , |
---|---|
Format: | text |
Language: | English |
Published: |
Institutional Knowledge at Singapore Management University
2021
|
Subjects: | |
Online Access: | https://ink.library.smu.edu.sg/sis_research/8155 https://ink.library.smu.edu.sg/context/sis_research/article/9158/viewcontent/Step_Wise_Deep_Learning_Models_for_Solving_Routing_Problems_av.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Singapore Management University |
Language: | English |
id |
sg-smu-ink.sis_research-9158 |
---|---|
record_format |
dspace |
spelling |
sg-smu-ink.sis_research-91582023-09-26T10:22:44Z Step-wise deep learning models for solving routing problems XIN, Liang SONG, Wen CAO, Zhiguang ZHANG, Jie Routing problems are very important in intelligent transportation systems. Recently, a number of deep learning-based methods are proposed to automatically learn construction heuristics for solving routing problems. However, these methods do not completely follow Bellman's Principle of Optimality since the visited nodes during construction are still included in the following subtasks, resulting in suboptimal policies. In this article, we propose a novel step-wise scheme which explicitly removes the visited nodes in each node selection step. We apply this scheme to two representative deep models for routing problems, pointer network and transformer attention model (TAM), and significantly improve the performance of the original models. To reduce computational complexity, we further propose the approximate step-wise TAM model by modifying one layer of attention. It enables training on larger instances compared to step-wise TAM, and outperforms state-of-the-art deep models with greedy decoding strategy. 2021-07-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/8155 info:doi/10.1109/TII.2020.3031409 https://ink.library.smu.edu.sg/context/sis_research/article/9158/viewcontent/Step_Wise_Deep_Learning_Models_for_Solving_Routing_Problems_av.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Routing Deep learning Decoding Computational modeling Reinforcement learning Urban areas Informatics Deep learning Deep reinforcement learning Intelligent transportation system Routing problems Numerical Analysis and Scientific Computing Operations Research, Systems Engineering and Industrial Engineering Transportation |
institution |
Singapore Management University |
building |
SMU Libraries |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
SMU Libraries |
collection |
InK@SMU |
language |
English |
topic |
Routing Deep learning Decoding Computational modeling Reinforcement learning Urban areas Informatics Deep learning Deep reinforcement learning Intelligent transportation system Routing problems Numerical Analysis and Scientific Computing Operations Research, Systems Engineering and Industrial Engineering Transportation |
spellingShingle |
Routing Deep learning Decoding Computational modeling Reinforcement learning Urban areas Informatics Deep learning Deep reinforcement learning Intelligent transportation system Routing problems Numerical Analysis and Scientific Computing Operations Research, Systems Engineering and Industrial Engineering Transportation XIN, Liang SONG, Wen CAO, Zhiguang ZHANG, Jie Step-wise deep learning models for solving routing problems |
description |
Routing problems are very important in intelligent transportation systems. Recently, a number of deep learning-based methods are proposed to automatically learn construction heuristics for solving routing problems. However, these methods do not completely follow Bellman's Principle of Optimality since the visited nodes during construction are still included in the following subtasks, resulting in suboptimal policies. In this article, we propose a novel step-wise scheme which explicitly removes the visited nodes in each node selection step. We apply this scheme to two representative deep models for routing problems, pointer network and transformer attention model (TAM), and significantly improve the performance of the original models. To reduce computational complexity, we further propose the approximate step-wise TAM model by modifying one layer of attention. It enables training on larger instances compared to step-wise TAM, and outperforms state-of-the-art deep models with greedy decoding strategy. |
format |
text |
author |
XIN, Liang SONG, Wen CAO, Zhiguang ZHANG, Jie |
author_facet |
XIN, Liang SONG, Wen CAO, Zhiguang ZHANG, Jie |
author_sort |
XIN, Liang |
title |
Step-wise deep learning models for solving routing problems |
title_short |
Step-wise deep learning models for solving routing problems |
title_full |
Step-wise deep learning models for solving routing problems |
title_fullStr |
Step-wise deep learning models for solving routing problems |
title_full_unstemmed |
Step-wise deep learning models for solving routing problems |
title_sort |
step-wise deep learning models for solving routing problems |
publisher |
Institutional Knowledge at Singapore Management University |
publishDate |
2021 |
url |
https://ink.library.smu.edu.sg/sis_research/8155 https://ink.library.smu.edu.sg/context/sis_research/article/9158/viewcontent/Step_Wise_Deep_Learning_Models_for_Solving_Routing_Problems_av.pdf |
_version_ |
1779157185550352384 |