ECML PKDD 2024 | 时空数据（Spatial-Temporal）和时间序列（Time series）论文总结

ECML PKDD 2024于9月9号-9月13号在立陶宛维尔纽斯举行（Vilnius）

本文总结了ECML PKDD 2024有关时空数据（spatial-temporal data）的相关论文，主要包含交通预测，预训练，迁移学习等内容，如有疏漏，欢迎大家补充。以及时间序列（time series）,包括时序预测，异常检测，分类，聚类等内容。

🌟【紧跟前沿】“时空探索之旅”与你一起探索时空奥秘！🚀
欢迎大家关注时空探索之旅时空探索之旅在这里插入图片描述

Research Track

时空：1-6 时序：7-15

1. Spatiotemporal Covariance Neural Networks

链接：https://link.springer.com/chapter/10.1007/978-3-031-70344-7_2

作者：Andrea Cavallo (Delft University of Technology)*; Mohammad Sabbaqi (Delft University of Technology); Elvin Isufi (Tu Delft)

关键词：多元时间序列，在线学习，PCA

2. Multivariate Traffic Demand Prediction via 2D Spectral Learning and Global Spatial Optimization

链接：https://link.springer.com/chapter/10.1007/978-3-031-70344-7_5

作者：Changlu Chen (UTS)*; Yanbin Liu (Auckland University of Technology); Ling Chen (" University of Technology, Sydney, Australia"); Chengqi Zhang (University of Technology Sydney)

关键词：交通需求预测，空间优化

3. Physics-Informed Spatio-Temporal Model for Human Mobility Prediction

链接：https://link.springer.com/chapter/10.1007/978-3-031-70344-7_24

作者：Quanyan Gao (Zhejiang University); Chao Li (Zhejiang University)*; Qinmin Yang (Zhejiang University)

关键词：人类移动性预测

4. Interpretable and Generalizable Spatiotemporal Predictive Learning with Disentangled Consistency

链接：https://link.springer.com/chapter/10.1007/978-3-031-70352-2_1

作者：Jingxuan Wei (Shenyang institute of computing technology, Chinese academy of sciences; University of Chinese Academy of Sciences)*; Cheng Tan (Zhejiang University & Westlake University); Zhangyang Gao (westlake university); Linzhuang Sun (Shenyang institute of computing technology, Chinese academy of sciences; University of Chinese Academy of Sciences); BiHui Yu (Shenyang institute of computing technology, Chinese academy of sciences); Ruifeng Guo (Shenyang institute of computing technology, Chinese academy of sciences); Stan Z. Li (Westlake University)

关键词：可解性，解耦，时空预测（更广义的）

5. Frequency Enhanced Pre-training for Cross-city Few-shot Traffic Forecasting

链接：https://link.springer.com/chapter/10.1007/978-3-031-70344-7_3

代码：https://github.com/zhyliu00/FEPCross

作者：Zhanyu Liu (Shanghai Jiao Tong University); Jianrong Ding (Shanghai Jiao Tong University); Guanjie Zheng (Shanghai Jiao Tong University)*

关键词：交通预测，预训练，少样本，跨城市迁移

FEPCross

6. Reinventing Node-Centric Traffic Forecasting for Improved Accuracy and Efficiency

链接：https://link.springer.com/chapter/10.1007/978-3-031-70352-2_2

作者：Xu Liu (National University of Singapore)*; Yuxuan Liang (The Hong Kong University of Science and Technology (Guangzhou)); Chao Huang (University of Hong Kong); Hengchang Hu (National University of Singapore); Yushi Cao (Nanyang Technological University); Bryan Hooi (National University of Singapore); Roger Zimmermann (NUS)

关键词：交通预测，预训练，少样本，跨城市迁移

7. Adaptive Sparsity Level during Training for Efficient Time Series Forecasting with Transformers

链接：https://link.springer.com/chapter/10.1007/978-3-031-70341-6_1

作者：Zahra Atashgahi (University of Twente)*; Mykola Pechenizkiy (TU Eindhoven); Raymond Veldhuis (University of Twente); Decebal Constantin Mocanu (University of Luxembourg)

关键词：时序预测，高效，稀疏性

8. Adaptive Seasonal-Trend Decomposition for Streaming Time Series Data with Transitions and Fluctuations in Seasonality

链接：https://link.springer.com/chapter/10.1007/978-3-031-70344-7_25

代码：https://sites.google.com/view/astd-ecmlpkdd/

作者：Thanapol Phungtua-eng (Shizuoka University)*; Yoshitaka Yamamoto

关键词：时序分解，流式数据

9. Diffusion model in Normal Gathering Latent Space for Time Series Anomaly Detection

链接：https://link.springer.com/chapter/10.1007/978-3-031-70352-2_17

作者：Jiashu Han (Harbin Institute of Technology); Shanshan Feng (Centre for Frontier AI Research, ASTAR); Min Zhou (Huawei Technologies co. ltd); Xinyu Zhang (Harbin Institute of Technology Shenzhen); Xutao Li (Harbin Institute of Technology Shenzhen Graduate School); Yew Soon Ong (Nanyang Technological University, Nanyang View, Singapore)

关键词：异常检测，隐扩散模型

10. Permutation Dependent Feature Mixing in TSMixer for Multivariate Time Series Forecasting

链接：https://link.springer.com/chapter/10.1007/978-3-031-70352-2_18

作者：rikuto yamazono (Wakayama University)*; Hirotaka Hachiya (Graduate School of System Engineering, Wakayama University)

关键词：时序预测（多元）

11. MMDL-based Data Augmentation with Domain Knowledge for Time Series Classification

链接：https://link.springer.com/chapter/10.1007/978-3-031-70352-2_24

作者：Xiaosheng Li (Ant Group); Yifan Wu (Peking University)*; Wei Jiang (Ant Group); Ying Li (Peking University); Jianguo Li (Ant Group)

关键词：时序分类，数据增强，领域知识

12. Improving the Evaluation and Actionability of Explanation Methods for Multivariate Time Series Classification

链接：https://link.springer.com/chapter/10.1007/978-3-031-70359-1_11

代码：https://github.com/mlgig/xai4mtsc_eval_actionability

作者：Davide Italo DI Serramazza (University College Dublin)*; Thach Le Nguyen (University College Dublin); Georgiana Ifrim (University College Dublin)

关键词：时序分类，可解释性，评测

13. Functional Latent Dynamics for Irregularly Sampled Time Series Forecasting

链接：https://link.springer.com/chapter/10.1007/978-3-031-70359-1_25

作者：Christian Klötergens (Information Science and Machine Learning Lab University of Hildesheim)*; Vijaya Yalavarthi (Information Systems and Machine Learning Lab, University of Hildesheim); Maximilian Stubbemann (Information Systems and Machine Learning Lab, University of Hildesheim); Lars Schmidt-Thieme (Universität Hildesheim)

关键词：不规则采样的时序预测，常微分方程

14. Graphical Model-Based Lasso for Weakly Dependent Time Series of Tensors

链接：https://link.springer.com/chapter/10.1007/978-3-031-70362-1_15

作者：Dorcas Ofori-Boateng (Portland State University)*; Jaidev Goel (The University Of Texas at Dallas); Yulia R. Gel (The University of Texas at Dallas); Ivor Cribben (University of Alberta)

关键词：图模型，lasso

15. Self-supervised Temporal and Spatial Normality Learning for Time Series Anomaly Detection

链接：https://link.springer.com/chapter/10.1007/978-3-031-70365-2_9

代码：https://github.com/mala-lab/STEN

作者：Yutong Chen (Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences)*; Hongzuo Xu (Intelligent Game and Decision Lab (IGDL)); Guansong Pang (Singapore Management University); Hezhe Qiao (Singapore Managment University); Yuan Zhou (Artificial Intelligence Research Center, DII); Mingsheng Shang (Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences)

关键词：异常检测，自监督，时空正态性

STEN

Applied Data Science Track

时空：16，17 时序：18，19

16. Spatial-Temporal PDE Networks for Traffic Flow Forecasting

链接：https://link.springer.com/chapter/10.1007/978-3-031-70381-2_11

作者：Tianshu Bao (Vanderbilt University)*, Hua Wei (Arizona State University), Junyi Ji (Vanderbilt University), Daniel Work (Vanderbilt University), Taylor T Johnson (Vanderbilt University)

关键词：交通预测，PDE

17. Spatial Transfer Learning for Estimating PM 2.5 in Data-poor Regions

链接：https://link.springer.com/chapter/10.1007/978-3-031-70378-2_24

作者：Shrey Gupta (Emory University)*, Yongbee Park (Inkgle), Jianzhao Bi (University of Washington), Suyash Gupta (University of California, Berkeley), Andreas Züfle (Emory University), Avani Wildani (Emory University), Yang Liu (Emory University)

关键词：PM2.5估计，迁移学习

18. Time Series Clustering for Enhanced Dynamic Allocation in A/B Testing

链接：https://link.springer.com/chapter/10.1007/978-3-031-70378-2_22

作者：Emmanuelle Claeys (IRIT)*, Myriam Maumy (UTT), Pierre Gançarski (University of Strasbourg)

关键词：时序聚类，A/B Testing

19. ExTea: An Evolutionary Algorithm-Based Approach for Enhancing Explainability in Time-Series Models

链接：https://link.springer.com/chapter/10.1007/978-3-031-70381-2_27

作者：Yiran Huang (Karlsruhe Institute of Technology)*, Yexu Zhou (KIT), Haibin Zhao (Karlsruhe Institute of Technology), Likun Fang (Karlsruhe Institute of Technology), Till Riedel (Karlsruhe Institute of Technology), Michael Beigl (Karlsruhe Institute of Technology)

关键词：可解释性

Demo Track

20. CityFlowER: An Efficient and Realistic Traffic Simulator with Embedded Machine Learning Models

链接：https://link.springer.com/chapter/10.1007/978-3-031-70371-3_22

代码：https://github.com/cityflow-project/CityFlowER

作者：Longchao Da, Chen Chu, Weinan Zhang, Hua Wei

关键词：交通模拟