{"type":"ScholarlyArticle","title":"Statistically-Guided Deep Network Transformation to Harness Heterogeneity in Space (Extended Abstract)","authors":[{"name":"Yiqun Xie","slug":"yiqun-xie","given":"Yiqun","family":"Xie","affiliation":"University of Maryland"},{"name":"Erhu He","slug":"erhu-he","given":"Erhu","family":"He","affiliation":"University of Pittsburgh"},{"name":"Xiaowei Jia","slug":"xiaowei-jia","given":"Xiaowei","family":"Jia","affiliation":"University of Pittsburgh"},{"name":"Han Bao","slug":"han-bao","given":"Han","family":"Bao","affiliation":"University of Iowa"},{"name":"Xun Zhou","slug":"xun-zhou","given":"Xun","family":"Zhou","affiliation":"University of Iowa"},{"name":"Rahul Ghosh","slug":"rahul-ghosh","given":"Rahul","family":"Ghosh","affiliation":"University of Minnesota"},{"name":"Praveen Ravirathinam","slug":"praveen-ravirathinam","given":"Praveen","family":"Ravirathinam","affiliation":"University of Minnesota"}],"abstract":"Spatial data are ubiquitous and have transformed decision-making in many critical domains, including public health, agriculture, transportation, etc. While recent advances in machine learning offer promising ways to harness massive spatial datasets (e.g., satellite imagery), spatial heterogeneity -- a fundamental property of spatial data -- poses a major challenge as data distributions or generative processes often vary over space. Recent studies targeting this difficult problem either require a known space-partitioning as the input, or can only support limited special cases (e.g., binary classification). Moreover, heterogeneity-pattern learned by these methods are locked to the locations of the training samples, and cannot be applied to new locations. We propose a statistically-guided framework to adaptively partition data in space during training using distribution-driven optimization and transform a deep learning model (of user's choice) into a heterogeneity-aware architecture. We also propose a spatial moderator to generalize learned patterns to new test regions. Experiment results on real-world datasets show that the framework can effectively capture footprints of heterogeneity and substantially improve prediction performances.","doi":"10.24963/ijcai.2022/752","url":"https://ijcai.org/proceedings/2022/752","pdf":"https://ijcai.org/proceedings/2022/0752.pdf","markdown":"https://ijcai.org/proceedings/2022/752.md","bibtex":"https://ijcai.org/proceedings/2022/bibtex/752","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":"Sister Conferences Best Papers","track":"Sister Conferences Best Papers","number":752,"pages":{"first":5364,"last":5368},"keywords":["Artificial Intelligence: General"],"videos":[{"label":"Video #1","url":"https://slideslive.com/38987262","length_sec":87,"provider":"slideslive"},{"label":"Video #2","url":"https://slideslive.com/38985405","length_sec":695,"provider":"slideslive"}],"references":[{"position":1,"raw":"[Atluri et al., 2018] Gowtham Atluri, Anuj Karpatne, and Vipin Kumar. Spatio-temporal data mining: A survey of problems and methods. ACM Computing Surveys (CSUR), 51(4):1-41, 2018.","title":"Spatio-temporal data mining: A survey of problems and methods","authors":"Gowtham Atluri, Anuj Karpatne, Vipin Kumar","year":"2018","venue":"ACM Computing Surveys (CSUR)","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Bao et al., 2020] Han Bao, Xun Zhou, Yingxue Zhang, Yanhua Li, and Yiqun Xie. Covid-gan: Estimating human mobility responses to covid-19 pandemic through spatio- temporal conditional generative adversarial networks. In Proceedings of the 28th international conference on ad- vances in geographic information systems, pages 273- 282, 2020.","title":"Covid-gan: Estimating human mobility responses to covid-19 pandemic through spatiotemporal conditional generative adversarial networks","authors":"Han Bao, Xun Zhou, Yingxue Zhang, Yanhua Li, Yiqun Xie","year":"2020","venue":"Proceedings of the 28th international conference on advances in geographic information systems","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Brunsdon et al., 1999] Chris Brunsdon, A Stewart Fother- ingham, and Martin Charlton. Some notes on parametric significance tests for geographically weighted regression. Journal of regional science, 39(3):497-524, 1999.","title":"Some notes on parametric significance tests for geographically weighted regression","authors":"Chris Brunsdon, Martin Stewart Fotheringham, Charlton","year":"1999","venue":"Journal of regional science","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[CDL, 2021] USDA cropland data layer. https://www.nass. usda.gov/Research and Science/Cropland/SARS1a.php, 2021. Accessed: 2022-06-03.","title":"USDA cropland data layer","authors":"Cdl","year":"2021","venue":null,"doi":null,"url":"https://www.nass.usda.gov/ResearchandScience/Cropland/SARS1a.php","internal_url":null},{"position":5,"raw":"[Finn et al., 2017] Chelsea Finn, Pieter Abbeel, and Sergey Levine. Model-agnostic meta-learning for fast adaptation of deep networks. In ICML, 2017.","title":"Model-agnostic meta-learning for fast adaptation of deep networks","authors":"Finn","year":"2017","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Fotheringham et al., 2017] A Stewart Fotheringham, Wen- bai Yang, and Wei Kang. Multiscale geographically weighted regression (mgwr). Annals of the American As- sociation of Geographers, 107(6):1247-1265, 2017.","title":"Multiscale geographically weighted regression (mgwr)","authors":"Wenbai Stewart Fotheringham, Wei Yang, Kang","year":"2017","venue":"Annals of the American Association of Geographers","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[GEOGLAM, 2021] Group on earth observations global agricultural monitoring initiative. https: //earthobservations.org/geoglam.php, 2021. Accessed: 2022-06-03.","title":"Group on earth observations global agricultural monitoring initiative","authors":"Geoglam","year":"2021","venue":null,"doi":null,"url":"https://earthobservations.org/geoglam.php","internal_url":null},{"position":8,"raw":"[Goodchild and Li, 2021] Michael F Goodchild and Wen- wen Li. Replication across space and time must be weak in the social and environmental sciences. Proceedings of the National Academy of Sciences, 118(35), 2021.","title":"Replication across space and time must be weak in the social and environmental sciences","authors":"F Michael, Wenwen Goodchild, Li","year":"2021","venue":"Proceedings of the National Academy of Sciences","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Gupta and others, 2020] Jayant Gupta et al. Towards spatial variability aware deep neural networks (svann): A sum- mary of results. In ACM SIGKDD workshop on deep learning for spatiotemporal data, app. & sys., 2020.","title":"Towards spatial variability aware deep neural networks (svann): A summary of results","authors":"Jayant Gupta","year":"2020","venue":"ACM SIGKDD workshop on deep learning for spatiotemporal data","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Gupta and others, 2021] Jayant Gupta et al. Spatial vari- ability aware deep neural networks (svann): A general approach. ACM Transactions on Intelligent Systems and Technology (TIST), 12(6):1-21, 2021.","title":"Spatial variability aware deep neural networks (svann): A general approach","authors":"Jayant Gupta","year":"2021","venue":"ACM Transactions on Intelligent Systems and Technology (TIST)","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Jia et al., 2019] Xiaowei Jia, Sheng Li, Ankush Khandel- wal, Guruprasad Nayak, Anuj Karpatne, and Vipin Kumar. Spatial context-aware networks for mining temporal dis- criminative period in land cover detection. In SDM, pages 513-521. SIAM, 2019.","title":"Spatial context-aware networks for mining temporal discriminative period in land cover detection","authors":"Xiaowei Jia, Sheng Li, Ankush Khandelwal, Guruprasad Nayak, Anuj Karpatne, Vipin Kumar","year":"2019","venue":"SDM","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Jiang et al., 2019] Zhe Jiang, Arpan Man Sainju, Yan Li, Shashi Shekhar, and Joseph Knight. Spatial ensemble learning for heterogeneous geographic data with class am- biguity. ACM Transactions on Intelligent Systems and Technology (TIST), 10(4):1-25, 2019.","title":"Spatial ensemble learning for heterogeneous geographic data with class ambiguity","authors":"Zhe Jiang, Arpan, Man Sainju, Yan Li, Shashi Shekhar, Joseph Knight","year":"2019","venue":"ACM Transactions on Intelligent Systems and Technology (TIST)","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Kraemer and others, 2020] Moritz UG Kraemer et al. The effect of human mobility and control measures on the covid-19 epidemic in china. Science, 368(6490):493-497, 2020.","title":"The effect of human mobility and control measures on the covid-19 epidemic in china","authors":"Kraemer, U G Moritz, Kraemer","year":"2020","venue":"Science","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Kratsios and Bilokopytov, 2020] Anastasis Kratsios and Ievgen Bilokopytov. Non-euclidean universal approxima- tion. NeurIPS, 33, 2020.","title":"Anastasis Kratsios and Ievgen Bilokopytov. Non-euclidean universal approximation","authors":"Bilokopytov Kratsios","year":"2020","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Krizhevsky et al., 2012] Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. Imagenet classification with deep convolutional neural networks. Advances in neural infor- mation processing systems, 25:1097-1105, 2012.","title":"Imagenet classification with deep convolutional neural networks","authors":"Alex Krizhevsky, Ilya Sutskever, Geoffrey E Hinton","year":"2012","venue":"Advances in neural information processing systems","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Kulldorff and others, 2007] Martin Kulldorff et al. Multi- variate scan statistics for disease surveillance. Statistics in medicine, 26(8):1824-1833, 2007.","title":"Multivariate scan statistics for disease surveillance","authors":"Martin Kulldorff, Kulldorff","year":"2007","venue":"Statistics in medicine","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Neill and others, 2013] Daniel B Neill et al. Fast subset scan for multivariate event detection. Statistics in medicine, 32(13):2185-2208, 2013.","title":"Fast subset scan for multivariate event detection","authors":"B Daniel, Neill","year":"2013","venue":"Statistics in medicine","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Neill, 2012] Daniel B Neill. Fast subset scan for spatial pat- tern detection. Journal of the Royal Statistical Society, 74(2):337-360, 2012.","title":"Fast subset scan for spatial pattern detection","authors":"B Daniel, Neill","year":"2012","venue":"Journal of the Royal Statistical Society","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Shekhar et al., 2015] Shashi Shekhar, Steven K Feiner, and Walid G Aref. Spatial computing. Communications of the ACM, 59(1):72-81, 2015.","title":"Spatial computing","authors":"Shashi Shekhar, Steven K Feiner, Walid G Aref","year":"2015","venue":"Communications of the ACM","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Xie et al., 2021a] Yiqun Xie, Erhu He, Xiaowei Jia, Han Bao, Xun Zhou, Rahul Ghosh, and Praveen Ravirathi- nam. A statistically-guided deep network transformation and moderation framework for data with spatial hetero- geneity. In 2021 IEEE International Conference on Data Mining (ICDM), pages 767-776. IEEE, 2021.","title":"A statistically-guided deep network transformation and moderation framework for data with spatial heterogeneity","authors":"Xie","year":"2021","venue":"IEEE International Conference on Data Mining (ICDM)","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Xie et al., 2021b] Yiqun Xie, Xiaowei Jia, Han Bao, Xun Zhou, Jia Yu, Rahul Ghosh, and Praveen Ravirathinam. Spatial-net: A self-adaptive and model-agnostic deep learning framework for spatially heterogeneous datasets. In Proceedings of the 29th International Conference on Advances in Geographic Information Systems, pages 313- 323, 2021.","title":"Spatial-net: A self-adaptive and model-agnostic deep learning framework for spatially heterogeneous datasets","authors":"Xie","year":"2021","venue":"Proceedings of the 29th International Conference on Advances in Geographic Information Systems","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Xie et al., 2021c] Yiqun Xie, Shashi Shekhar, and Yan Li. Statistically-robust clustering techniques for mapping spa- tial hotspots: A survey. ACM Computing Surveys (CSUR), 2021.","title":"Statistically-robust clustering techniques for mapping spatial hotspots: A survey","authors":"Xie","year":"2021","venue":"ACM Computing Surveys (CSUR)","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Yao et al., 2019] Huaxiu Yao, Yiding Liu, Ying Wei, Xian- feng Tang, and Zhenhui Li. Learning from multiple cities: A meta-learning approach for spatial-temporal prediction. In The World Wide Web Conference, pages 2181-2191, 2019.","title":"Learning from multiple cities: A meta-learning approach for spatial-temporal prediction","authors":"Huaxiu Yao, Yiding Liu, Ying Wei, Xianfeng Tang, Zhenhui Li","year":"2019","venue":"The World Wide Web Conference","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}