STEMO: Early Spatio-temporal Forecasting with Multi-Objective Reinforcement Learning

被引:0
作者
Shao, Wei [1 ]
Kang, Yufan [2 ]
Peng, Ziyan [3 ]
Xiao, Xiao [3 ]
Wang, Lei [4 ]
Yang, Yuhui [3 ]
Salim, Flora D. [5 ]
机构
[1] CSIRO, Data61, Clayton, Vic, Australia
[2] RMIT Univ, Melbourne, Vic, Australia
[3] Xidian Univ, Xian, Peoples R China
[4] Zhejiang Univ, Hangzhou, Peoples R China
[5] Univ New South Wales, Sydney, Australia
来源
PROCEEDINGS OF THE 30TH ACM SIGKDD CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, KDD 2024 | 2024年
基金
中国国家自然科学基金; 澳大利亚研究理事会;
关键词
Spatio-Temporal Data; Graph Neural Network; Early Detection; Responsible AI;
D O I
10.1145/3637528.3671922
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Accuracy and timeliness are indeed often conflicting goals in prediction tasks. Premature predictions may yield a higher rate of false alarms, whereas delaying predictions to gather more information can render them too late to be useful. In applications such as wildfires, crimes, and traffic jams, timely forecasting are vital for safeguarding human life and property. Consequently, finding a balance between accuracy and timeliness is crucial. In this paper, we propose an early spatio-temporal forecasting model based on Multi-Objective reinforcement learning that can either implement an optimal policy given a preference or infer the preference based on a small number of samples. The model addresses two primary challenges: 1) enhancing the accuracy of early forecasting and 2) providing the optimal policy for determining the most suitable prediction time for each area. Our method demonstrates superior performance on three large-scale real-world datasets, surpassing existing methods in early spatio-temporal forecasting tasks.
引用
收藏
页码:2618 / 2627
页数:10
相关论文
共 38 条
[1]  
Ahmed MS., 1979, TRANSP RES REC, V722, P1
[2]  
[Anonymous], 2016, MACHINE LEARNING KNO
[3]   STConvS2S: Spatiotemporal Convolutional Sequence to Sequence Network for weather forecasting [J].
Castro, Rafaela ;
Souto, Yania M. ;
Ogasawara, Eduardo ;
Porto, Fabio ;
Bezerra, Eduardo .
NEUROCOMPUTING, 2021, 426 :285-298
[4]  
Cho K., 2014, P SSST 8 8 WORKSH SY, DOI DOI 10.3115/V1/W14-4012
[5]   Emerging Trends Word2Vec [J].
Church, Kenneth Ward .
NATURAL LANGUAGE ENGINEERING, 2017, 23 (01) :155-162
[6]  
Deb K., 2002, Running Performance Metrics for Evolutionary Multi-Objective Optimization
[7]  
Nguyen D, 2018, IEEE INT CONF BIG DA, P4045, DOI 10.1109/BigData.2018.8621872
[8]   Spatiotemporal traffic forecasting: review and proposed directions [J].
Ermagun, Alireza ;
Levinson, David .
TRANSPORT REVIEWS, 2018, 38 (06) :786-814
[9]  
Ghalwash M.F., 2012, Bioinformatics and Biomedicine (BIBM), 2012 IEEE International Conference on, P1
[10]   Utilizing Temporal Patterns for Estimating Uncertainty in Interpretable Early Decision Making [J].
Ghalwash, Mohamed F. ;
Radosavljevic, Vladan ;
Obradovic, Zoran .
PROCEEDINGS OF THE 20TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING (KDD'14), 2014, :402-411