{"type":"ScholarlyArticle","title":"Physics-Informed Long-Sequence Forecasting From Multi-Resolution Spatiotemporal Data","authors":[{"name":"Chuizheng Meng","slug":"chuizheng-meng","given":"Chuizheng","family":"Meng","affiliation":"University of Southern California"},{"name":"Hao Niu","slug":"hao-niu","given":"Hao","family":"Niu","affiliation":"KDDI Research , Inc."},{"name":"Guillaume Habault","slug":"guillaume-habault","given":"Guillaume","family":"Habault","affiliation":"KDDI Research , Inc."},{"name":"Roberto Legaspi","slug":"roberto-legaspi","given":"Roberto","family":"Legaspi","affiliation":"KDDI Research , Inc."},{"name":"Shinya Wada","slug":"shinya-wada","given":"Shinya","family":"Wada","affiliation":"KDDI Research , Inc."},{"name":"Chihiro Ono","slug":"chihiro-ono","given":"Chihiro","family":"Ono","affiliation":"KDDI Research , Inc."},{"name":"Yan Liu","slug":"yan-liu","given":"Yan","family":"Liu","affiliation":"University of Southern California"}],"abstract":"Spatiotemporal data aggregated over regions or time windows at various resolutions demonstrate heterogeneous patterns and dynamics in each resolution. Meanwhile, the multi-resolution characteristic provides rich contextual information, which is critical for effective long-sequence forecasting. The importance of such inter-resolution information is more significant in practical cases, where fine-grained data is usually collected via approaches with lower costs but also lower qualities compared to those for coarse-grained data. However, existing works focus on uni-resolution data and cannot be directly applied to fully utilize the aforementioned extra information in multi-resolution data. In this work, we propose Spatiotemporal Koopman Multi-Resolution Network (ST-KMRN), a physics-informed learning framework for long-sequence forecasting from multi-resolution spatiotemporal data. Our method jointly models data aggregated in multiple resolutions and captures the inter-resolution dynamics with the self-attention mechanism. We also propose downsampling and upsampling modules among resolutions to further strengthen the connections among data of multiple resolutions. Moreover, we enhance the modeling of intra-resolution dynamics with physics-informed modules based on Koopman theory. Experimental results demonstrate that our proposed approach achieves the best performance on the long-sequence forecasting tasks compared to baselines without a specific design for multi-resolution data.","doi":"10.24963/ijcai.2022/304","url":"https://ijcai.org/proceedings/2022/304","pdf":"https://ijcai.org/proceedings/2022/0304.pdf","markdown":"https://ijcai.org/proceedings/2022/304.md","bibtex":"https://ijcai.org/proceedings/2022/bibtex/304","volume":{"slug":"2022","title":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2022","url":"https://ijcai.org/proceedings/2022/","year":2022,"location":"Vienna","dates":"23-29 July 2022","editors":"Luc De Raedt","isbn":"978-1-956792-00-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2022"},"section":"Data Mining","track":"Main Track","number":304,"pages":{"first":2189,"last":2195},"keywords":["Machine Learning: Sequence and Graph Learning","Data Mining: Mining Spatial and/or Temporal Data"],"videos":[{"label":"Video #1","url":"https://slideslive.com/38986991","length_sec":85,"provider":"slideslive"},{"label":"Video #2","url":"https://slideslive.com/38985134","length_sec":683,"provider":"slideslive"}],"references":[{"position":1,"raw":"Obs Ratio 0.8 0.6 0.4 0.2 MAE RMSE MAE RMSE MAE RMSE MAE RMSE HA 21.95 35.52 21.97 35.58 21.99 35.62 22.08 35.82 Static 14.21 26.66 15.77 29.73 22.49 40.69 41.19 61.80 GRU 22.83(1.15) 34.97(1.62) 27.74(0.40) 41.21(0.46) 27.75(0.40) 41.27(0.45) 27.78(0.42) 41.31(0.46) Informer 19.04(0.98) 29.21(1.29) 23.83(1.93) 35.35(2.50) 22.90(0.35) 34.31(0.50) 23.30(1.53) 34.66(1.96) Graph WaveNet 16.53(0.29) 31.10(0.43) 17.08(0.30) 31.01(0.39) 16.74(0.21) 30.76(0.35) 16.75(0.33) 30.72(0.16) MTGNN 18.43(0.88) 32.73(2.02) 19.41(0.40) 34.88(0.71) 19.81(0.09) 35.49(0.11) 18.95(0.11) 33.46(0.66) KoopmanAE 18.43(1.57) 29.79(2.20) 19.90(1.47) 31.58(1.94) 18.96(2.08) 30.09(2.99) 19.07(0.58) 30.66(0.77) ST-KMRN 12.56(0.66) 22.58(0.71) 11.66(0.36) 21.46(0.58) 11.38(0.28) 21.17(0.45) 11.76(0.31) 21.64(0.48) RelErr -11.6% -15.3% -26.1% -27.8% -32.0% -29.6% -29.8% -29.4% RelErrGW -24.0% -27.4% -31.7% -30.8% -32.0% -31.2% -29.8% -29.6%","title":".8 0.6 0.4 0","authors":null,"year":null,"venue":"MAE RMSE MAE RMSE MAE RMSE MAE RMSE HA","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Azencot et al., 2020] Omri Azencot, N Benjamin Erichson, Vanessa Lin, and Michael Mahoney. Forecasting sequential data using consistent koopman autoencoders. In Interna- tional Conference on Machine Learning, pages 475-485.","title":"Forecasting sequential data using consistent koopman autoencoders","authors":"Omri Azencot, N Benjamin Erichson, Vanessa Lin, Michael Mahoney","year":null,"venue":"International Conference on Machine Learning","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"PMLR, 2020.","title":null,"authors":"Pmlr","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Bai et al., 2020] Lei Bai, Lina Yao, Can Li, Xianzhi Wang, and Can Wang. Adaptive graph convolutional recurrent network for traffic forecasting. Advances in Neural Infor- mation Processing Systems, 33, 2020.","title":"Adaptive graph convolutional recurrent network for traffic forecasting","authors":"Lei Bai, Lina Yao, Can Li, Xianzhi Wang, Can Wang","year":"2020","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Cao et al., 2020] Defu Cao, Yujing Wang, Juanyong Duan, Ce Zhang, Xia Zhu, Congrui Huang, Yunhai Tong, Bixiong Xu, Jing Bai, Jie Tong, and Qi Zhang. Spectral temporal graph neural network for multivariate time-series forecast- ing. In Advances in Neural Information Processing Systems (NeurIPS) 33, pages 17766-17778, 2020.","title":"Spectral temporal graph neural network for multivariate time-series forecasting","authors":"Defu Cao, Yujing Wang, Juanyong Duan, Ce Zhang, Xia Zhu, Congrui Huang, Yunhai Tong, Bixiong Xu, Jing Bai, Jie Tong, Qi Zhang","year":"2020","venue":"Advances in Neural Information Processing Systems (NeurIPS)","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Chen et al., 2020] Weiqi Chen, Ling Chen, Yu Xie, Wei Cao, Yusong Gao, and Xiaojie Feng. Multi-range attentive bi- component graph convolutional network for traffic forecast- ing. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pages 3529-3536, 2020.","title":"Multi-range attentive bicomponent graph convolutional network for traffic forecasting","authors":"Weiqi Chen, Ling Chen, Yu Xie, Wei Cao, Yusong Gao, Xiaojie Feng","year":"2020","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Chung et al., 2014] Junyoung Chung, Caglar Gulcehre, KyungHyun Cho, and Yoshua Bengio. Empirical evaluation of gated recurrent neural networks on sequence modeling. arXiv preprint arXiv:1412.3555, 2014.","title":"Empirical evaluation of gated recurrent neural networks on sequence modeling","authors":"Chung","year":"2014","venue":null,"doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Geng et al., 2019] Xu Geng, Yaguang Li, Leye Wang, Lingyu Zhang, Qiang Yang, Jieping Ye, and Yan Liu. Spatiotemporal multi-graph convolution network for ride- hailing demand forecasting. In Proceedings of the AAAI conference on artificial intelligence, volume 33, pages 3656-3663, 2019.","title":"Spatiotemporal multi-graph convolution network for ridehailing demand forecasting","authors":"Yaguang Xu Geng, Leye Li, Lingyu Wang, Qiang Zhang, Jieping Yang, Yan Ye, Liu","year":"2019","venue":"Proceedings of the AAAI conference on artificial intelligence","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Kipf and Welling, 2017] Thomas N Kipf and Max Welling. Semi-supervised classification with graph convolutional networks. In ICLR, 2017.","title":"Semi-supervised classification with graph convolutional networks","authors":"N Thomas, Max Kipf, Welling","year":"2017","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Koopman, 1931] Bernard O Koopman. Hamiltonian systems and transformation in hilbert space. Proceedings of the national academy of sciences of the united states of america, 17(5):315, 1931.","title":"Hamiltonian systems and transformation in hilbert space","authors":"Bernard O Koopman","year":"1931","venue":"Proceedings of the national academy of sciences of the united states of america","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Li et al., 2021] Fuxian Li, Jie Feng, Huan Yan, Guangyin Jin, Fan Yang, Funing Sun, Depeng Jin, and Yong Li. Dynamic graph convolutional recurrent network for traffic prediction: Benchmark and solution. ACM Transactions on Knowledge Discovery from Data (TKDD), 2021. [NREL, 2021] NREL. https://www.nrel.gov/grid/ solar-power-data.html, 2021. Accessed: 2021-10-01.","title":"Dynamic graph convolutional recurrent network for traffic prediction: Benchmark and solution","authors":"Li","year":"2021","venue":"ACM Transactions on Knowledge Discovery from Data (TKDD)","doi":null,"url":"https://www.nrel.gov/grid/solar-power-data.html,2021","internal_url":null},{"position":12,"raw":"[NYCTLC, 2021] NYCTLC. https://www1.nyc.gov/site/tlc/ about/tlc-trip-record-data.page, 2021. Accessed: 2021-10- 01.","title":"NYCTLC","authors":"Nyctlc","year":"2021","venue":null,"doi":null,"url":"https://www1.nyc.gov/site/tlc/about/tlc-trip-record-data.page","internal_url":null},{"position":13,"raw":"[Oord et al., 2016] Aaron van den Oord, Sander Dieleman, Heiga Zen, Karen Simonyan, Oriol Vinyals, Alex Graves, Nal Kalchbrenner, Andrew Senior, and Koray Kavukcuoglu. Wavenet: A generative model for raw audio. arXiv preprint arXiv:1609.03499, 2016.","title":"Wavenet: A generative model for raw audio","authors":"Oord","year":"2016","venue":null,"doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Shi et al., 2016] Wenzhe Shi, Jose Caballero, Ferenc Huszár, Johannes Totz, Andrew P Aitken, Rob Bishop, Daniel Rueckert, and Zehan Wang. Real-time single image and video super-resolution using an efficient sub-pixel convo- lutional neural network. In Proceedings of the IEEE con- ference on computer vision and pattern recognition, pages 1874-1883, 2016.","title":"Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network","authors":"Shi","year":"2016","venue":"Proceedings of the IEEE conference on computer vision and pattern recognition","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Vaswani et al., 2017] Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need. In NIPS, 2017.","title":"Attention is all you need","authors":"Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Lukasz Kaiser, Illia Polosukhin","year":"2017","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Wu et al., 2019] Zonghan Wu, Shirui Pan, Guodong Long, Jing Jiang, and Chengqi Zhang. Graph wavenet for deep spatial-temporal graph modeling. In IJCAI, 2019.","title":"Graph wavenet for deep spatial-temporal graph modeling","authors":"Wu","year":"2019","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2019/264"},{"position":17,"raw":"[Wu et al., 2020] Zonghan Wu, Shirui Pan, Guodong Long, Jing Jiang, Xiaojun Chang, and Chengqi Zhang. Connect- ing the dots: Multivariate time series forecasting with graph neural networks. In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD '20, page 753-763, New York, NY, USA, 2020. Association for Computing Machinery.","title":"Connecting the dots: Multivariate time series forecasting with graph neural networks","authors":"Wu","year":"2020","venue":"Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD '20","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Yao et al., 2019] Huaxiu Yao, Xianfeng Tang, Hua Wei, Guanjie Zheng, and Zhenhui Li. Revisiting spatial-temporal similarity: A deep learning framework for traffic predic- tion. In Proceedings of the AAAI conference on artificial intelligence, volume 33, pages 5668-5675, 2019.","title":"Revisiting spatial-temporal similarity: A deep learning framework for traffic prediction","authors":"Yao","year":"2019","venue":"Proceedings of the AAAI conference on artificial intelligence","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Zhang et al., 2017] Junbo Zhang, Yu Zheng, and Dekang Qi. Deep spatio-temporal residual networks for citywide crowd flows prediction. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 31, 2017.","title":"Deep spatio-temporal residual networks for citywide crowd flows prediction","authors":"Junbo Zhang, Yu Zheng, Dekang Qi","year":"2017","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Zheng et al., 2020] Chuanpan Zheng, Xiaoliang Fan, Cheng Wang, and Jianzhong Qi. Gman: A graph multi-attention network for traffic prediction. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pages 1234-1241, 2020.","title":"Gman: A graph multi-attention network for traffic prediction","authors":"Zheng","year":"2020","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Zhou et al., 2020] Zhengyang Zhou, Yang Wang, Xike Xie, Lianliang Chen, and Hengchang Liu. Riskoracle: A minute- level citywide traffic accident forecasting framework. In Proceedings of the AAAI Conference on Artificial Intelli- gence, volume 34, pages 1258-1265, 2020.","title":"Riskoracle: A minutelevel citywide traffic accident forecasting framework","authors":"Zhou","year":"2020","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Zhou et al., 2021] Haoyi Zhou, Shanghang Zhang, Jieqi Peng, Shuai Zhang, Jianxin Li, Hui Xiong, and Wancai Zhang. Informer: Beyond efficient transformer for long sequence time-series forecasting. In The Thirty-Fifth AAAI Conference on Artificial Intelligence, AAAI 2021, page on- line. AAAI Press, 2021.","title":"Informer: Beyond efficient transformer for long sequence time-series forecasting","authors":"Zhou","year":"2021","venue":"The Thirty-Fifth AAAI Conference on Artificial Intelligence, AAAI 2021, page online","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}