AAAI 2024 | 时间序列（Time Series）和时空数据（Spatial-Temporal）论文总结

时空探索之旅

发布于 2024-11-19 16:08:23

8720

文章被收录于专栏：时空探索之旅时空探索之旅

AAAI今年共有12100篇投稿（Main Technical Track），有9862篇经过严格审稿，共录取了2342篇论文，录取率23.75%。。

AAAI 2024将在2024年2月22日到25日于加拿大温哥华举行。

本文总结了2024 AAAI上有关时空数据（spatial-temporal）时间序列（time series）数据相关论文。

时空数据Topic:交通预测，轨迹表示学习，信控优化等

时间序列Topic：时间序列预测，分类，异常检测，因果发现等

时间序列（time series）

1. MSGNet: Learning Multi-Scale Inter-Series Correlations for Multivariate Time Series Forecasting

作者：Cai, Wanlin; Liang, Yuxuan; LIU, XIANGGEN; Feng, Jianshuai; Wu, Yuankai*

关键词：多元时间序列预测，相关性

arXiv：https://arxiv.org/abs/2401.00423

解读1：多元时序预测：多尺度下的多变量关系学习

解读2：AAAI 2024 | MSGNet：学习多尺度序列之间的相关性以进行多元时间序列预测

2. Learning from Polar Representation: An Extreme-Adaptive Model for Long-Term Time Series Forecasting

作者：Li, Yanhong; Xu, Jack; Anastasiu, David C*

关键词：长时预测

arXiv：https://arxiv.org/abs/2312.08763

DAN

3. Graph-Aware Contrasting for Multivariate Time-Series Classification

作者：Wang, Yucheng*; Xu, Yuecong; Yang, Jianfei; Wu, Min; Li, Xiaoli; Xie, Lihua; Chen, Zhenghua

关键词：多元时间序列分类，图对比

arXiv：https://arxiv.org/abs/2309.05202v3

TS-GAC

4. U-Mixer: An Unet-Mixer Architecture with Stationarity Correction for Time Series Forecasting

作者：Ma, Xiang; Li, Xuemei; Fang, Lexin; Zhao, Tianlong; Zhang, Caiming*

关键词：平稳性校正，时间序列预测

arXiv：https://arxiv.org/abs/2401.02236

U-Mixer

5. [Oral] GraFITi: Graphs for Forecasting Irregularly Sampled Time Series

作者：Yalavarthi, Vijaya Krishna*; Madhusudhanan, Kiran; Scholz, Randolf; Ahmed, Nourhan; Burchert, Johannes; Jawed, Shayan; Born, Stefan; Schmidt-Thieme, Lars

关键词：不规则时间序列，图（Graph）

arXiv链接：https://arxiv.org/abs/2305.12932

GraFITi

6. IVP-VAE: Modeling EHR Time Series with Initial Value Problem Solvers

作者：Xiao, Jingge*; Basso, Leonie; Nejdl, Wolfgang; Ganguly, Niloy; Sikdar, Sandipan

关键词：时间序列建模（EHR）

arXiv链接：https://arxiv.org/abs/2305.06741

Modeling irregular time series with IVP-VAE

7. Cross-Domain Contrastive Learning for Time Series Clustering

作者：Peng, Furong*; jike, luo; Lu, Xuan; Wang, Sheng; Li, Feijiang

关键词：跨域对比学习，时间序列聚类

CDCC

8. SimPSI: A Simple Strategy to Preserve Spectral Information in Time Series Data Augmentation

作者：Ryu, Hyun*; Yoon, Sunjae; Yoon, Hee Suk; Yoon, Eunseop; Yoo, Chang D.

关键词：谱域、时间序列数据增强

Code：https://github.com/Hyun-Ryu/simpsi

arXiv：https://arxiv.org/abs/2312.05790

SimPSI

9. TimesURL: Self-supervised Contrastive Learning for Universal Time Series Representation Learning

作者：Liu, jiexi*; Chen, Songcan

关键词：自监督，对比学习，时间序列表示学习

arXiv：https://arxiv.org/abs/2312.15709

解读：TimesURL：探索时间序列的表示学习

TimesURL

10. CGS-Mask: Making Time Series Predictions Intuitive for All

作者：Lu, Feng; Li, Wei*; Sun, Yifei; Song, Cheng; Yufei, Ren; Zomaya, Albert

关键词：时间序列预测，可解释性（intuitive）

arXiv：https://arxiv.org/abs/2312.09513

CGS-Mask

11. Diffusion Language-Shapelets for Semisupervised Time-series Classification

作者：Liu, Zhen; Pei, Wenbin; Lan, Disen; Ma, Qianli*

关键词：半监督，Shapelet，时间序列分类

12. CUTS+: High-dimensional Causal Discovery from Irregular Time-series

作者：Yuxiao Cheng, Lianglong Li, Tingxiong Xiao, Zongren Li, Qin Zhong, Jinli Suo, Kunlun He

关键词：因果发现，不规则时间序列

Code：https://github.com/jarrycyx/unn

arXiv：https://arxiv.org/abs/2305.05890

CUTS+

13. When Model Meets New Normals: Test-time Adaptation for Unsupervised Time-series Anomaly Detection

作者：Kim, Dongmin*; Park, Sunghyun ; Choo, Jaegul

关键词：无监督，Test-time Adaptation（TTA），时间序列异常检测

Code：https://github.com/carrtesy/M2N2

arXiv：https://arxiv.org/abs/2312.11976

Illustration on detrend module

14. HDMixer: Hierarchical Dependency with Extendable Patch for Multivariate Time Series Forecasting

作者：Huang, Qihe*; Shen, Lei; Zhang, Ruixin; Cheng, Jiahuan; Ding, Shouhong; Zhou, Zhengyang ; Wang, Yang

关键词：多层级，多元时间序列预测

15. Energy-efficient Streaming Time Series Classification with Attentive Power Iteration

作者：Huang, Hao*; Shah, Tapan; Evans, Scott; Yoo, Shinjae

关键词：时间序列分类

16. Latent Diffusion Transformer for Probabilistic Time Series Forecasting

作者：Feng, Shibo*; Miao, Chunyan; Zhang, Zhong; Zhao, Peilin

关键词：扩散Transfomer，概率时间预测

17. Considering Nonstationary within Multivariate Time Series with Variational Hierarchical Transformer for Forecasting

作者：Wang, Muyao*; Chen, Wenchao; Chen, Bo

关键词：非平稳性，多层级，多元时间序列

时空数据（spatial-temporal data）

1. Spatial-Temporal Interplay in Human Mobility: A Hierarchical Reinforcement Learning Approach with Hypergraph Representation

作者：Zhang, Zhaofan; Xiao, Yanan; Jiang, Lu; Yang, Dingqi; Yin, Minghao; Wang, Pengyang

关键词：人类移动性，超图，多层次，强化学习

arXiv：https://arxiv.org/abs/2312.15717

STI-HRL

2. Fully-Connected Spatial-Temporal Graph for Multivariate Time Series Data

作者：Wang, Yucheng*; Xu, Yuecong; Yang, Jianfei; Wu, Min; Li, Xiaoli; Xie, Lihua; Chen, Zhenghua

关键词：时空图，多元时间序列

arXiv：https://arxiv.org/abs/2309.05305

解读：AAAI 2024 | 多维时序下的全连接时空图模型

FC-STGNN

3. Earthfarsser: Versatile Spatio-Temporal Dynamical Systems Modeling in One Model

作者：Wu, Hao*; Liang, Yuxuan; Xiong, Wei; Zhou, Zhengyang ; Huang, Wei; wang, shilong; wang, kun

关键词：时空动态系统，气象预测

Code：https://github.com/easylearningscores/EarthFarseer

arXiv：https://arxiv.org/abs/2312.08403

Earthfarsser

4. [Oral] Prompt to transfer: Sim-to-real Transfer for Traffic Signal Control with Prompt Learning

作者：Da, Longchao; Gao, Mingquan; Wei, Hua*; Da, Longchao; mei, hao

关键词：提示学习，信控优化，sim2real

arXiv：https://arxiv.org/abs/2308.14284

PromptGAT

5. Urban Region Embedding via Multi-View Contrastive Prediction

作者：Li, Zechen; Huang, Weiming; Zhao, Kai; Yang, Min; Gong, Yongshun; Chen, Meng*

关键词：表示学习，对比学习

ReCP

6. KGTS: Contrastive Trajectory Similarity Learning over Prompt Knowledge Graph Embedding

作者：Chen, Zhen; Zhang, Dalin; Feng, Shanshan; Chen, Kaixuan; Chen, Lisi; Han, Peng; Shang, Shuo*

关键词：对比学习，轨迹相似度，知识图谱，表示学习

7. CI-STHPAN: Pre-Trained Attention Network for Stock Selection with Channel-Independent Spatio-Temporal Hypergraph

作者：Xia, Hongjie; Ao, Huijie; Li, Long; Liu, Yu; Liu, Sen; Ye, Guangnan*; Chai, Hongfeng

关键词：预训练，通道独立（CI）,时空超图

8. Hawkes-enhanced Spatial-Temporal Hypergraph Contrastive Learning based on Criminal Correlations

作者：Liang, Ke*; Zhou, Sihang; Liu, Meng; Liu, Yue; Tu, Wenxuan; Zhang, Yi; Fang, Liming; Liu, Zhe; Liu, Xinwang

关键词：犯罪预测，时空超图，对比学习

9. ModWaveMLP: MLP-based Mode Decomposition and Wavelet Denoising Model to Defeat Complex Structures in Traffic Forecasting

作者：Sun, Ke*; Liu, Pei; Li, Pengfei; Liao, Zhifang

关键词：交通预测，小波去噪， MLP

10. Spatio-Temporal Pivotal Graph Neural Networks for Traffic Flow Forecasting

作者：Kong, Weiyang; Guo, Ziyu; Liu, Yubao

关键词：交通预测，时空图神经网络

11. Towards Streaming Spatial-Temporal Graph Learning: A Decoupled Perspective

作者：Wang, Binwu; Wang, Pengkun; Zhang, Yudong; Wang, Xu; Zhou, Zhengyang ; Bai, Lei; Wang, Yang

关键词：时空图，解耦

12. Successive POI Recommendation via Brain-inspired Spatiotemporal Aware Representation

作者：Ma, Gehua ; Wang, He; Zhao, Jingyuan; Yan, Rui; Tang, Huajin*

关键词：POI推荐，脑启发式，时空感知建模（表示）

13. Learning Time Slot Preferences via Mobility Tree for Next POI Recommendation

作者：Huang, Tianhao Alex; Pan, Xuan; Cai, Xiangrui*; ZHANG, Ying; Yuan, Xiaojie

关键词：POI推荐

Code：https://github.com/Skyyyy0920/MTNet