A fast matrix autoregression algorithm based on Tucker decomposition for online prediction of nonlinear real-time taxi-hailing demand without pre-training

IF 5.3 1区数学 Q1 MATHEMATICS, INTERDISCIPLINARY APPLICATIONS

Chaos Solitons & Fractals Pub Date : 2024-10-16 DOI:10.1016/j.chaos.2024.115660

Zhihao Xu, Zhiqiang Lv, Benjia Chu, Jianbo Li

{"title":"A fast matrix autoregression algorithm based on Tucker decomposition for online prediction of nonlinear real-time taxi-hailing demand without pre-training","authors":"Zhihao Xu, Zhiqiang Lv, Benjia Chu, Jianbo Li","doi":"10.1016/j.chaos.2024.115660","DOIUrl":null,"url":null,"abstract":"<div><div>Online prediction of real-time taxi-hailing demand generally provides better real-time decision support for passengers and taxi drivers compared with offline prediction. Current studies focused on using deep spatial-temporal models to predict complex nonlinear taxi-hailing demand. However, whether these models can be used for online prediction of real-time taxi-hailing demand through online training or offline pre-training is hardly discussed. Generally, deep models are not lightweight enough for online training, and pre-training these models requires some time and computational resources. Therefore, a lightweight Fast Matrix Autoregression algorithm based on Tucker Decomposition (FMAR-TD) is proposed for online real-time training and prediction of nonlinear taxi-hailing demand without pre-training. The experimental results show that FMAR-TD achieves millisecond-level online prediction of real-time taxi-hailing demand. Compared with baselines, the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) of FMAR-TD marginally increase by 2.51 % and 2.56 %, while the computation time (sum of training time and prediction time) significantly reduces by 86.16 %. Open-source link: <span><span>https://github.com/qdu318/FMAR-TD</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":9764,"journal":{"name":"Chaos Solitons & Fractals","volume":"189 ","pages":"Article 115660"},"PeriodicalIF":5.3000,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Chaos Solitons & Fractals","FirstCategoryId":"100","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0960077924012128","RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MATHEMATICS, INTERDISCIPLINARY APPLICATIONS","Score":null,"Total":0}

引用次数: 0

Abstract

Online prediction of real-time taxi-hailing demand generally provides better real-time decision support for passengers and taxi drivers compared with offline prediction. Current studies focused on using deep spatial-temporal models to predict complex nonlinear taxi-hailing demand. However, whether these models can be used for online prediction of real-time taxi-hailing demand through online training or offline pre-training is hardly discussed. Generally, deep models are not lightweight enough for online training, and pre-training these models requires some time and computational resources. Therefore, a lightweight Fast Matrix Autoregression algorithm based on Tucker Decomposition (FMAR-TD) is proposed for online real-time training and prediction of nonlinear taxi-hailing demand without pre-training. The experimental results show that FMAR-TD achieves millisecond-level online prediction of real-time taxi-hailing demand. Compared with baselines, the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) of FMAR-TD marginally increase by 2.51 % and 2.56 %, while the computation time (sum of training time and prediction time) significantly reduces by 86.16 %. Open-source link: https://github.com/qdu318/FMAR-TD.

查看原文本刊更多论文

基于塔克分解的快速矩阵自回归算法，无需预训练即可在线预测非线性实时打车需求

与离线预测相比，实时打车需求的在线预测通常能为乘客和出租车司机提供更好的实时决策支持。目前的研究侧重于使用深度时空模型来预测复杂的非线性打车需求。然而，这些模型是否可以通过在线训练或离线预训练用于实时打车需求的在线预测，目前几乎没有讨论。一般来说，深度模型不够轻量级，不适合在线训练，而且预训练这些模型需要一定的时间和计算资源。因此，本文提出了一种基于塔克分解的轻量级快速矩阵自回归算法（FMAR-TD），用于非线性打车需求的在线实时训练和预测，无需预训练。实验结果表明，FMAR-TD 实现了毫秒级的实时打车需求在线预测。与基线相比，FMAR-TD 的平均绝对误差（MAE）和均方根误差（RMSE）略微增加了 2.51 % 和 2.56 %，而计算时间（训练时间与预测时间之和）则大幅减少了 86.16 %。开源链接：https://github.com/qdu318/FMAR-TD。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Chaos Solitons & Fractals 物理-数学跨学科应用

CiteScore

13.20

自引率

10.30%

发文量

1087

审稿时长

9 months

期刊介绍： Chaos, Solitons & Fractals strives to establish itself as a premier journal in the interdisciplinary realm of Nonlinear Science, Non-equilibrium, and Complex Phenomena. It welcomes submissions covering a broad spectrum of topics within this field, including dynamics, non-equilibrium processes in physics, chemistry, and geophysics, complex matter and networks, mathematical models, computational biology, applications to quantum and mesoscopic phenomena, fluctuations and random processes, self-organization, and social phenomena.