{"type":"ScholarlyArticle","title":"When Transfer Learning Meets Cross-City Urban Flow Prediction: Spatio-Temporal Adaptation Matters","authors":[{"name":"Ziquan Fang","slug":"ziquan-fang","given":"Ziquan","family":"Fang","affiliation":"College of Computer Science , Zhejiang University , Hangzhou , China,"},{"name":"Dongen Wu","slug":"dongen-wu","given":"Dongen","family":"Wu","affiliation":"College of Computer Science , Zhejiang University of Technology , Hangzhou , China"},{"name":"Lu Pan","slug":"lu-pan","given":"Lu","family":"Pan","affiliation":"College of Computer Science , Zhejiang University , Hangzhou , China,"},{"name":"Lu Chen","slug":"lu-chen","given":"Lu","family":"Chen","affiliation":"College of Computer Science , Zhejiang University , Hangzhou , China,"},{"name":"Yunjun Gao","slug":"yunjun-gao","given":"Yunjun","family":"Gao","affiliation":"College of Computer Science , Zhejiang University , Hangzhou , China,"}],"abstract":"Urban flow prediction is a fundamental task to build smart cities, where neural networks have become the most popular method. However, the deep learning methods typically rely on massive training data that are probably inaccessible in real world. In light of this, the community calls for knowledge transfer. However, when adapting transfer learning for cross-city prediction tasks, existing studies are built on static knowledge transfer, ignoring the fact inter-city correlations change dynamically across time. The dynamic correlations make urban feature transfer challenging. This paper proposes a novel Spatio-Temporal Adaptation Network (STAN) to perform urban flow prediction for data-scarce cities via the spatio-temporal knowledge transferred from data-rich cities. STAN encompasses three modules: i) spatial adversarial adaptation module that adopts an adversarial manner to capture the transferable spatial features; ii) temporal attentive adaptation module to attend to critical dynamics for temporal feature transfer; iii) prediction module that aims to learn task-driven transferable knowledge. Extensive experiments on five real datasets show STAN substantially outperforms state-of-the-art methods.","doi":"10.24963/ijcai.2022/282","url":"https://ijcai.org/proceedings/2022/282","pdf":"https://ijcai.org/proceedings/2022/0282.pdf","markdown":"https://ijcai.org/proceedings/2022/282.md","bibtex":"https://ijcai.org/proceedings/2022/bibtex/282","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":282,"pages":{"first":2030,"last":2036},"keywords":["Data Mining: Mining Spatial and/or Temporal Data"],"videos":[{"label":"Video #1","url":"https://slideslive.com/38986714","length_sec":89,"provider":"slideslive"},{"label":"Video #2","url":"https://slideslive.com/38984857","length_sec":293,"provider":"slideslive"}],"references":[{"position":1,"raw":"[Bai et al., 2020] Lei Bai, Lina Yao, Can Li, Xianzhi Wang, and Can Wang. Adaptive graph convolutional recurrent network for traffic forecasting. In NeurIPS, 2020.","title":"Adaptive graph convolutional recurrent network for traffic forecasting","authors":"Lei Bai, Lina Yao, Can Li, Xianzhi Wang, Can Wang","year":"2020","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Chen et al., 2019] Min-Hung Chen, Zsolt Kira, Ghassan Al- regib, Jaekwon Yoo, Ruxin Chen, and Jian Zheng. Tempo- ral attentive alignment for large-scale video domain adap- tation. In ICCV, pages 6320-6329, 2019.","title":"Temporal attentive alignment for large-scale video domain adaptation","authors":"Chen","year":"2019","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Fang et al., 2021] Ziquan Fang, Lu Pan, Lu Chen, Yuntao Du, and Yunjun Gao. Mdtp: A multi-source deep traffic prediction framework over spatio-temporal trajectory data. VLDB, 14(8):1289-1297, 2021.","title":"Mdtp: A multi-source deep traffic prediction framework over spatio-temporal trajectory data","authors":"Fang","year":"2021","venue":"VLDB","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Goodfellow et al., 2014] Ian J. Goodfellow, Jean Pouget- Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron C. Courville, and Yoshua Bengio. Generative adversarial nets. In NeurIPS, 2014.","title":"Generative adversarial nets","authors":"Ian J Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron C Courville, Yoshua Bengio","year":"2014","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Guo et al., 2021] Shengnan Guo, Youfang Lin, Huaiyu Wan, Xiucheng Li, and Gao Cong. Learning dynamics and heterogeneity of spatial-temporal graph data for traf- fic forecasting. TKDE, 2021.","title":"Learning dynamics and heterogeneity of spatial-temporal graph data for traffic forecasting","authors":"Guo","year":"2021","venue":"TKDE","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Jiang et al., 2021] Renhe Jiang, Du Yin, Zhaonan Wang, Yizhuo Wang, Jiewen Deng, Hangchen Liu, Zekun Cai, Jinliang Deng, Xuan Song, and Ryosuke Shibasaki. Dl- traff: Survey and benchmark of deep learning models for urban traffic prediction. In CIKM, pages 4515-4525, 2021.","title":"Dltraff: Survey and benchmark of deep learning models for urban traffic prediction","authors":"Renhe Jiang, Du Yin, Zhaonan Wang, Yizhuo Wang, Jiewen Deng, Hangchen Liu, Zekun Cai, Jinliang Deng, Xuan Song, Ryosuke Shibasaki","year":"2021","venue":"CIKM","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Li et al., 2018a] Yaguang Li, Rose Yu, Cyrus Shahabi, and Yan Liu. Diffusion convolutional recurrent neural net- work: Data-driven traffic forecasting. In ICLR, 2018.","title":"Diffusion convolutional recurrent neural network: Data-driven traffic forecasting","authors":"Li","year":"2018","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Li et al., 2018b] Yaguang Li, Rose Yu, Cyrus Shahabi, and Yan Liu. Diffusion convolutional recurrent neural net- work: Data-driven traffic forecasting. In ICLR, 2018.","title":"Diffusion convolutional recurrent neural network: Data-driven traffic forecasting","authors":"Li","year":"2018","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Liang et al., 2019] Yuxuan Liang, Kun Ouyang, Lin Jing, Sijie Ruan, Ye Liu, Junbo Zhang, David S. Rosenblum, and Yu Zheng. Urbanfm: Inferring fine-grained urban flows. In KDD, pages 3132-3142, 2019.","title":"Urbanfm: Inferring fine-grained urban flows","authors":"Yuxuan Liang, Kun Ouyang, Lin Jing, Sijie Ruan, Ye Liu, Junbo Zhang, David S Rosenblum, Yu Zheng","year":"2019","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Liao et al., 2018] Binbing Liao, Jingqing Zhang, Chao Wu, Douglas McIlwraith, Tong Chen, Shengwen Yang, Yike Guo, and Fei Wu. Sequence learning with auxiliary for traffic prediction. In KDD, pages 537-546, 2018.","title":"Sequence learning with auxiliary for traffic prediction","authors":"Liao","year":"2018","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Long et al., 2015] Mingsheng Long, Yue Cao, Jianmin Wang, and Michael Jordan. Learning transferable features with adaptation networks. In ICML, pages 97-105, 2015.","title":"Learning transferable features with adaptation networks","authors":"Mingsheng Long, Yue Cao, Jianmin Wang, Michael Jordan","year":"2015","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Long et al., 2017] Mingsheng Long, Han Zhu, Jianmin Wang, and Michael I Jordan. Deep transfer learning with joint adaptation networks. In ICML, pages 2208-2217, 2017.","title":"Deep transfer learning with joint adaptation networks","authors":"Mingsheng Long, Han Zhu, Jianmin Wang, Michael I Jordan","year":"2017","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Mazimpaka and Timpf, 2016] Jean Damascène Mazimpaka and Sabine Timpf. Trajectory data mining: A review of methods and applications. Journal of Spatial Information Science, 2016(13):61-99, 2016.","title":"Jean Damascène Mazimpaka and Sabine Timpf. Trajectory data mining: A review of methods and applications","authors":"Timpf Mazimpaka","year":"2016","venue":"Journal of Spatial Information Science","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Pei et al., 2018] Zhongyi Pei, Zhangjie Cao, Mingsheng Long, and Jianmin Wang. Multi-adversarial domain adap- tation. In AAAI, 2018.","title":"Multi-adversarial domain adaptation","authors":"Zhongyi Pei, Zhangjie Cao, Mingsheng Long, Jianmin Wang","year":"2018","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Tong et al., 2017] Yongxin Tong, Lei Chen, and Cyrus Sha- habi. Spatial crowdsourcing: Challenges, techniques, and applications. VLDB, 10(12):1988-1991, 2017.","title":"Spatial crowdsourcing: Challenges, techniques, and applications","authors":"Tong","year":"2017","venue":"VLDB","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Wang et al., 2018] Leye Wang, Bin Guo, and Qiang Yang. Smart city development with urban transfer learning. Computer, 51(12):32-41, 2018.","title":"Smart city development with urban transfer learning","authors":"Leye Wang, Bin Guo, Qiang Yang","year":"2018","venue":"Computer","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Wang et al., 2019] Leye Wang, Xu Geng, Xiaojuan Ma, Feng Liu, and Qiang Yang. Cross-city transfer learning for deep spatio-temporal prediction. In IJCAI, 2019.","title":"Cross-city transfer learning for deep spatio-temporal prediction","authors":"Leye Wang, Xu Geng, Xiaojuan Ma, Feng Liu, Qiang Yang","year":"2019","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2019/262"},{"position":18,"raw":"[Wang et al., 2021] Senzhang Wang, Hao Miao, Jiyue Li, and Jiannong Cao. Spatio-temporal knowledge transfer for urban crowd flow prediction via deep attentive adaptation networks. TITS, 2021. [Xingjian et al., 2015] SHI Xingjian, Zhourong Chen, Hao Wang, Dit-Yan Yeung, Wai-Kin Wong, and Wang-chun Woo. Convolutional lstm network: A machine learning approach for precipitation nowcasting. In NeurIPS, 2015.","title":"Spatio-temporal knowledge transfer for urban crowd flow prediction via deep attentive adaptation networks","authors":"Wang","year":"2015","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Yao et al., 2019a] 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 WWW, pages 2181-2191, 2019.","title":"Learning from multiple cities: A meta-learning approach for spatial-temporal prediction","authors":"Yao","year":"2019","venue":"WWW","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Yao et al., 2019b] Huaxiu Yao, Xianfeng Tang, Hua Wei, Guanjie Zheng, and Zhenhui Li. Revisiting spatial- temporal similarity: A deep learning framework for traffic prediction. In AAAI, pages 5668-5675, 2019.","title":"Revisiting spatialtemporal similarity: A deep learning framework for traffic prediction","authors":"Yao","year":"2019","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Ye et al., 2019] Junchen Ye, Leilei Sun, Bowen Du, Yanjie Fu, Xinran Tong, and Hui Xiong. Co-prediction of multi- ple transportation demands based on deep spatio-temporal neural network. In KDD, pages 305-313, 2019.","title":"Co-prediction of multiple transportation demands based on deep spatio-temporal neural network","authors":"Ye","year":"2019","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Ye et al., 2020] Jiexia Ye, Juanjuan Zhao, Kejiang Ye, and Chengzhong Xu. How to build a graph-based deep learn- ing architecture in traffic domain: A survey. TITS, 2020.","title":"How to build a graph-based deep learning architecture in traffic domain: A survey","authors":"Ye","year":"2020","venue":"TITS","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Yuan and Li, 2021] Haitao Yuan and Guoliang Li. A survey of traffic prediction: from spatio-temporal data to intelli- gent transportation. DSE, 6(1):63-85, 2021.","title":"A survey of traffic prediction: from spatio-temporal data to intelligent transportation","authors":"Haitao Yuan, Guoliang Li","year":"2021","venue":"DSE","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Yuan et al., 2020] Haitao Yuan, Guoliang Li, Zhifeng Bao, and Ling Feng. Effective travel time estimation: When historical trajectories over road networks matter. In SIG- MOD, pages 2135-2149, 2020.","title":"Effective travel time estimation: When historical trajectories over road networks matter","authors":"Haitao Yuan, Guoliang Li, Zhifeng Bao, Ling Feng","year":"2020","venue":"SIG-MOD","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Yuan et al., 2021] Haitao Yuan, Guoliang Li, Zhifeng Bao, and Ling Feng. Effective joint prediction model for travel demands and traffic flows. In ICDE, pages 348-359, 2021.","title":"Effective joint prediction model for travel demands and traffic flows","authors":"Haitao Yuan, Guoliang Li, Zhifeng Bao, Ling Feng","year":"2021","venue":"ICDE","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Zhang et al., 2016] Junbo Zhang, Yu Zheng, Dekang Qi, Ruiyuan Li, and Xiuwen Yi. Dnn-based prediction model for spatio-temporal data. In SIGSPATIAL, 2016.","title":"Dnn-based prediction model for spatio-temporal data","authors":"Junbo Zhang, Yu Zheng, Dekang Qi, Ruiyuan Li, Xiuwen Yi","year":"2016","venue":"SIGSPATIAL","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Zhang et al., 2017] Junbo Zhang, Yu Zheng, and Dekang Qi. Deep spatio-temporal residual networks for citywide crowd flows prediction. In AAAI, 2017.","title":"Deep spatio-temporal residual networks for citywide crowd flows prediction","authors":"Junbo Zhang, Yu Zheng, Dekang Qi","year":"2017","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Zhang et al., 2020] Junbo Zhang, Yu Zheng, Junkai Sun, and Dekang Qi. Flow prediction in spatio-temporal networks based on multitask deep learning. TKDE, 32(3):468-478, 2020.","title":"Flow prediction in spatio-temporal networks based on multitask deep learning","authors":"Junbo Zhang, Yu Zheng, Junkai Sun, Dekang Qi","year":"2020","venue":"TKDE","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Zheng, 2015] Yu Zheng. Trajectory data mining: an overview. TIST, 6(3):1-41, 2015.","title":"Trajectory data mining: an overview","authors":"Yu Zheng","year":"2015","venue":"TIST","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}