{"type":"ScholarlyArticle","title":"MUSCAT: Multi-Scale Spatio-Temporal Learning with Application to Climate Modeling","authors":[{"name":"Jianpeng Xu","slug":"jianpeng-xu","given":"Jianpeng","family":"Xu","affiliation":"Department of Computer Science and Engineering , Michigan State University"},{"name":"Xi Liu","slug":"xi-liu","given":"Xi","family":"Liu","affiliation":"Department of Computer Science and Engineering , Michigan State University"},{"name":"Tyler Wilson","slug":"tyler-wilson","given":"Tyler","family":"Wilson","affiliation":"Department of Computer Science and Engineering , Michigan State University"},{"name":"Pang-Ning Tan","slug":"pang-ning-tan","given":"Pang-Ning","family":"Tan","affiliation":"Department of Computer Science and Engineering , Michigan State University"},{"name":"Pouyan Hatami","slug":"pouyan-hatami","given":"Pouyan","family":"Hatami","affiliation":"Department of Geography , Michigan State University {xujianpe ,"},{"name":"Lifeng Luo","slug":"lifeng-luo","given":"Lifeng","family":"Luo","affiliation":"Department of Geography , Michigan State University {xujianpe ,"}],"abstract":"In climate and environmental sciences, vast amount of spatio-temporal data have been generated at varying spatial resolutions from satellite observations and computer models. Integrating such diverse sources of data has proven to be useful for building prediction models as the multi-scale data may capture different aspects of the Earth system. In this paper, we present a novel framework called MUSCAT for predictive modeling of multi-scale, spatio-temporal data. MUSCAT performs a joint decomposition of multiple tensors from different spatial scales, taking into account the relationships between the variables. The latent factors derived from the joint tensor decomposition are  used to train the spatial and temporal prediction models at different scales for each location. The outputs from these ensemble of spatial and temporal models will be aggregated to generate future predictions. An incremental learning algorithm is also proposed to handle the massive size of the tensors. Experimental results on real-world data from the United States Historical Climate Network (USHCN) showed that MUSCAT outperformed other competing methods in more than 70\\% of the locations.","doi":"10.24963/ijcai.2018/404","url":"https://ijcai.org/proceedings/2018/404","pdf":"https://ijcai.org/proceedings/2018/0404.pdf","markdown":"https://ijcai.org/proceedings/2018/404.md","bibtex":"https://ijcai.org/proceedings/2018/bibtex/404","volume":{"slug":"2018","title":"Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2018","url":"https://ijcai.org/proceedings/2018/","year":2018,"location":"Stockholm","dates":"13-19 July 2018","editors":"Jérôme Lang","isbn":"978-0-9992411-2-7","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2018"},"section":"Machine Learning","track":"Main track","number":404,"pages":{"first":2912,"last":2918},"keywords":["Machine Learning: Feature Selection ; Learning Sparse Models","Machine Learning: Transfer, Adaptation, Multi-task Learning"],"videos":[],"references":[{"position":1,"raw":"[Acar et al., 2011] Evrim Acar, Tamara G. Kolda, and Daniel M. Dunlavy. All-at-once optimization for coupled matrix and tensor factorizations. In MLG'11: Proceedings of Mining and Learning with Graphs, August 2011.","title":"All-at-once optimization for coupled matrix and tensor factorizations","authors":"Evrim Acar, Tamara G Kolda, Daniel M Dunlavy","year":"2011","venue":"MLG'11: Proceedings of Mining and Learning with Graphs","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Bellocchio et al., 2012] F. Bellocchio, S. Ferrari, V. Piuri, and N.A. Borghese. Hierarchical approach for multiscale support vector regression. IEEE TNNLS, 23(9):1448- 1460, September 2012.","title":"Hierarchical approach for multiscale support vector regression","authors":"F Bellocchio, S Ferrari, V Piuri, N A Borghese","year":"2012","venue":"IEEE TNNLS","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Bertasius et al., 2015] Gedas Bertasius, Jianbo Shi, and Lorenzo Torresani. Deepedge: A multi-scale bifurcated deep network for top-down contour detection. In CVPR, pages 4380-4389, 2015.","title":"Deepedge: A multi-scale bifurcated deep network for top-down contour detection","authors":"Gedas Bertasius, Jianbo Shi, Lorenzo Torresani","year":"2015","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Caruana, 1997] Rich Caruana. Multitask learning. Machine Learning, 28(1):41-75, July 1997.","title":"Multitask learning","authors":"Caruana","year":"1997","venue":"Machine Learning","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Eigen et al., 2014] David Eigen, Christian Puhrsch, and Rob Fergus. Depth map prediction from a single image using a multi-scale deep network. In NIPS, pages 2366-2374, 2014.","title":"Depth map prediction from a single image using a multi-scale deep network","authors":"David Eigen, Christian Puhrsch, Rob Fergus","year":"2014","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Ermiş et al., 2015] Beyza Ermiş, Evrim Acar, and A. Taylan Cemgil. Link prediction in heterogeneous data via general- ized coupled tensor factorization. Data Mining and Knowl- edge Discovery, 29(1):203-236, Jan 2015.","title":"Link prediction in heterogeneous data via generalized coupled tensor factorization","authors":"Beyza Ermiş, Evrim Acar, A Taylan Cemgil","year":"2015","venue":"Data Mining and Knowledge Discovery","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Gonçalves et al., 2016] André R. Gonçalves, Fernando J. Von Zuben, and Arindam Banerjee. Multi-task sparse structure learning with gaussian copula models. JMLR, 17(1):1205-1234, January 2016.","title":"Multi-task sparse structure learning with gaussian copula models","authors":"R André, Fernando J Gonçalves, Arindam Von Zuben, Banerjee","year":"2016","venue":"JMLR","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Gonçalves et al., 2017] André R. Gonçalves, Arindam Banerjee, and Fernando J. Von Zuben. Spatial projection of multiple climate variables using hierarchical multitask learning. In AAAI, pages 4509-4515, 2017.","title":"Spatial projection of multiple climate variables using hierarchical multitask learning","authors":"R André, Arindam Gonçalves, Fernando J Banerjee, Von Zuben","year":"2017","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Kolda and Bader, 2009] Tamara G. Kolda and Brett W. Bader. Tensor decompositions and applications. SIAM Re- view, 51(3):455-500, August 2009.","title":"Tensor decompositions and applications","authors":"Tamara G Kolda, Brett W Bader","year":"2009","venue":"SIAM Review","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Miller et al., 2015] Bradley A. Miller, Sylvia Koszinski, Marc Wehrhan, and Michael Sommer. Impact of multi- scale predictor selection for modeling soil properties. Geo- derma, 239-240:97 -106, 2015.","title":"Impact of multiscale predictor selection for modeling soil properties","authors":"Bradley A Miller, Sylvia Koszinski, Marc Wehrhan, Michael Sommer","year":"2015","venue":"Geoderma","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Neverova et al., 2014] Natalia Neverova, Christian Wolf, Graham W. Taylor, and Florian Nebout. Multi-scale deep learning for gesture detection and localization. In ECCV Workshops, pages 474-490, 2014.","title":"Multi-scale deep learning for gesture detection and localization","authors":"Natalia Neverova, Christian Wolf, Graham W Taylor, Florian Nebout","year":"2014","venue":"ECCV Workshops","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Nounou and Nounou, 2010] Mohamed N. Nounou and Hazem N. Nounou. Multiscale latent variable regression. International Journal of Chemical Engineering, 2010, 2010.","title":"Multiscale latent variable regression","authors":"Mohamed N Nounou, Hazem N Nounou","year":"2010","venue":"International Journal of Chemical Engineering","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Parikh and Boyd, 2014] Neal Parikh and Stephen Boyd. Proximal algorithms. Foundations and Trends in Opti- mization, 1(3):127-239, January 2014.","title":"Proximal algorithms","authors":"Neal Parikh, Stephen Boyd","year":"2014","venue":"Foundations and Trends in Optimization","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Steinbach et al., 2003] Michael Steinbach, Pang-Ning Tan, Vipin Kumar, Steven Klooster, and Christopher Potter. Discovery of climate indices using clustering. In KDD, pages 446-455, 2003.","title":"Discovery of climate indices using clustering","authors":"Michael Steinbach, Pang-Ning Tan, Vipin Kumar, Steven Klooster, Christopher Potter","year":"2003","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Sun et al., 2008] Jimeng Sun, Dacheng Tao, Spiros Pa- padimitriou, Philip S. Yu, and Christos Faloutsos. Incre- mental tensor analysis: Theory and applications. TKDD, 2(3):11:1-11:37, 2008.","title":"Incremental tensor analysis: Theory and applications","authors":"Sun","year":"2008","venue":"TKDD","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Walder et al., 2008] Christian Walder, Kwang In Kim, and Bernhard Schölkopf. Sparse multiscale gaussian process regression. In ICML, pages 1112-1119, 2008.","title":"Sparse multiscale gaussian process regression","authors":"Walder","year":"2008","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Xu et al., 2016a] Jianpeng Xu, Pang-Ning Tan, Lifeng Luo, and Jiayu Zhou. Gspartan: a geospatio-temporal multi-task learning framework for multi-location prediction. In SDM, pages 657-665, 2016.","title":"Gspartan: a geospatio-temporal multi-task learning framework for multi-location prediction","authors":"Xu","year":"2016","venue":"SDM","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Xu et al., 2016b] Jianpeng Xu, Jiayu Zhou, Pang-Ning Tan, Xi Liu, and Lifeng Luo. Wisdom: Weighted incremental spatio-temporal multi-task learning via tensor decomposi- tion. In IEEE Big Data, pages 522-531, Dec 2016.","title":"Wisdom: Weighted incremental spatio-temporal multi-task learning via tensor decomposition","authors":"Xu","year":"2016","venue":"IEEE Big Data","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Xu et al., 2017] Jianpeng Xu, Pang-Ning Tan, Jiayu Zhou, and Lifeng Luo. Online multi-task learning framework for ensemble forecasting. IEEE TKDE, 29(6):1268-1280, 2017.","title":"Online multi-task learning framework for ensemble forecasting","authors":"Xu","year":"2017","venue":"IEEE TKDE","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Yu et al., 2015] Rose Yu, Dehua Cheng, and Yan Liu. Ac- celerated online low rank tensor learning for multivariate spatiotemporal streams. In ICML, volume 37, pages 238- 247, 2015.","title":"Accelerated online low rank tensor learning for multivariate spatiotemporal streams","authors":"Rose Yu, Dehua Cheng, Yan Liu","year":"2015","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Zhao and Du, 2016] Wenzhi Zhao and Shihong Du. Learn- ing multiscale and deep representations for classifying re- motely sensed imagery. JPRS, 113:155 -165, 2016.","title":"Learning multiscale and deep representations for classifying remotely sensed imagery","authors":"Du Zhao, Wenzhi Zhao, Shihong Du","year":"2016","venue":"JPRS","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Zhao et al., 2015] Liang Zhao, Qian Sun, Jieping Ye, Feng Chen, Chang-Tien Lu, and Naren Ramakrishnan. Multi- task learning for spatio-temporal event forecasting. In KDD, pages 1503-1512, 2015.","title":"Multitask learning for spatio-temporal event forecasting","authors":"Liang Zhao, Qian Sun, Jieping Ye, Feng Chen, Chang-Tien Lu, Naren Ramakrishnan","year":"2015","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Zhou et al., 2016] Shuo Zhou, Xuan Vinh Nguyen, James Bailey, Yunzhe Jia, and Ian Davidson. Accelerating Online CP Decompositions for Higher Order Tensors. In KDD, pages 1375-1384, 2016.","title":"Accelerating Online CP Decompositions for Higher Order Tensors","authors":"Shuo Zhou, Xuan Vinh Nguyen, James Bailey, Yunzhe Jia, Ian Davidson","year":"2016","venue":"KDD","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}