{"type":"ScholarlyArticle","title":"Coherent Probabilistic Aggregate Queries on Long-horizon Forecasts","authors":[{"name":"Prathamesh Deshpande","slug":"prathamesh-deshpande","given":"Prathamesh","family":"Deshpande","affiliation":"Department of Computer Science and Engineering , IIT Bombay"},{"name":"Sunita Sarawagi","slug":"sunita-sarawagi","given":"Sunita","family":"Sarawagi","affiliation":"Department of Computer Science and Engineering , IIT Bombay"}],"abstract":"Long range forecasts are the starting point of many decision support systems that need to draw inference from high-level aggregate patterns on forecasted values. State of the art time-series forecasting methods are either subject to concept drift on long-horizon forecasts, or fail to accurately predict coherent and accurate high-level aggregates.\r\n\r\nIn this work, we present a novel probabilistic forecasting method that produces forecasts that are coherent in terms of base level and predicted aggregate statistics. We achieve the coherency between predicted base-level and aggregate statistics using a novel inference method based on KL-divergence that can be solved efficiently in closed form. We show that our method improves forecast performance across both base level and unseen aggregates post inference on real datasets ranging three diverse domains. (Project URL)","doi":"10.24963/ijcai.2022/404","url":"https://ijcai.org/proceedings/2022/404","pdf":"https://ijcai.org/proceedings/2022/0404.pdf","markdown":"https://ijcai.org/proceedings/2022/404.md","bibtex":"https://ijcai.org/proceedings/2022/bibtex/404","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":"Machine Learning","track":"Main Track","number":404,"pages":{"first":2916,"last":2922},"keywords":["Machine Learning: Probabilistic Machine Learning","Machine Learning: Multi-view learning","Machine Learning: Time-series; Data Streams","Data Mining: Mining Data Streams"],"videos":[{"label":"Video #1","url":"https://slideslive.com/38987112","length_sec":74,"provider":"slideslive"},{"label":"Video #2","url":"https://slideslive.com/38985255","length_sec":676,"provider":"slideslive"}],"references":[{"position":1,"raw":"[Bansal et al., 2021] Parikshit Bansal, Prathamesh Desh- pande, and Sunita Sarawagi. Missing value imputa- tion on multidimensional time series. arXiv preprint arXiv:2103.01600, 2021.","title":"Missing value imputation on multidimensional time series","authors":"Parikshit Bansal, Prathamesh Deshpande, Sunita Sarawagi","year":"2021","venue":null,"doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Ben Taieb and Koo, 2019] Souhaib Ben Taieb and Bonsoo Koo. Regularized regression for hierarchical forecasting without unbiasedness conditions. In Proceedings of the 25th ACM SIGKDD International Conference on Knowl- edge Discovery and Data Mining, 2019.","title":"Regularized regression for hierarchical forecasting without unbiasedness conditions","authors":"Ben Taieb, Koo ; ] Souhaib, Ben Taieb, Bonsoo Koo","year":"2019","venue":"Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Benidis et al., 2020] Konstantinos Benidis, Syama Sundar Rangapuram, Valentin Flunkert, Bernie Wang, Danielle Maddix, Caner Turkmen, Jan Gasthaus, Michael Bohlke- Schneider, David Salinas, Lorenzo Stella, et al. Neural forecasting: Introduction and literature overview. arXiv preprint arXiv:2004.10240, 2020.","title":"Neural forecasting: Introduction and literature overview","authors":"Konstantinos Benidis, Syama Sundar Rangapuram, Valentin Flunkert, Bernie Wang, Danielle Maddix, Caner Turkmen, Jan Gasthaus, Michael Bohlke-Schneider, David Salinas, Lorenzo Stella","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Deshpande and Sarawagi, 2019] Prathamesh Deshpande and Sunita Sarawagi. Streaming adaptation of deep forecasting models using adaptive recurrent units. In ACM SIGKDD, 2019.","title":"Streaming adaptation of deep forecasting models using adaptive recurrent units","authors":"Prathamesh Deshpande, Sunita Sarawagi","year":"2019","venue":"ACM SIGKDD","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Deshpande et al., 2021] Prathamesh Deshpande, Kamlesh Marathe, Abir De, and Sunita Sarawagi. Long hori- zon forecasting with temporal point processes. In ACM WSDM, 2021.","title":"Long horizon forecasting with temporal point processes","authors":"Prathamesh Deshpande, Kamlesh Marathe, Abir De, Sunita Sarawagi","year":"2021","venue":"ACM WSDM","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Flunkert et al., 2017] Valentin Flunkert, David Salinas, and Jan Gasthaus. Deepar: Probabilistic forecasting with au- toregressive recurrent networks. CoRR, abs/1704.04110, 2017.","title":"Deepar: Probabilistic forecasting with autoregressive recurrent networks","authors":"Valentin Flunkert, David Salinas, Jan Gasthaus","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Gasthaus et al., 2019] Jan Gasthaus, Konstantinos Benidis, Yuyang Wang, Syama Sundar Rangapuram, David Sali- nas, Valentin Flunkert, and Tim Januschowski. Probabilis- tic forecasting with spline quantile function rnns. In The 22nd International Conference on Artificial Intelligence and Statistics, pages 1901-1910, 2019.","title":"Probabilistic forecasting with spline quantile function rnns","authors":"Gasthaus","year":"2019","venue":"The 22nd International Conference on Artificial Intelligence and Statistics","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Guen and Thome, 2020] Vincent Le Guen and Nicolas Thome. Probabilistic time series forecasting with struc- tured shape and temporal diversity. arXiv preprint arXiv:2010.07349, 2020.","title":"Probabilistic time series forecasting with structured shape and temporal diversity","authors":"Thome ; Guen, Le Vincent, Nicolas Guen, Thome","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Han et al., 2021] Xing Han, Sambarta Dasgupta, and Joy- deep Ghosh. Simultaneously reconciled quantile forecast- ing of hierarchically related time series. In Proceedings of The 24th International Conference on Artificial Intelli- gence and Statistics, pages 190-198, 2021.","title":"Simultaneously reconciled quantile forecasting of hierarchically related time series","authors":"Xing Han, Sambarta Dasgupta, Joydeep Ghosh","year":"2021","venue":"Proceedings of The 24th International Conference on Artificial Intelligence and Statistics","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[LE GUEN and THOME, 2019] Vincent LE GUEN and Nicolas THOME. Shape and time distortion loss for train- ing deep time series forecasting models. In Advances in Neural Information Processing Systems 32. 2019.","title":"Shape and time distortion loss for training deep time series forecasting models","authors":"L E Guen, Thome ; Vincent, L E Guen, Nicolas Thome","year":"2019","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Li et al., 2019] Shiyang Li, Xiaoyong Jin, Yao Xuan, Xiyou Zhou, Wenhu Chen, Yu-Xiang Wang, and Xifeng Yan. En- hancing the locality and breaking the memory bottleneck of transformer on time series forecasting. In Advances in Neural Information Processing Systems, pages 5243- 5253, 2019.","title":"Enhancing the locality and breaking the memory bottleneck of transformer on time series forecasting","authors":"Shiyang Li, Xiaoyong Jin, Xiyou Yao Xuan, Wenhu Zhou, Yu-Xiang Chen, Xifeng Wang, Yan","year":"2019","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Lim et al., 2020] Bryan Lim, Sercan O. Arik, Nicolas Lo- eff, and Tomas Pfister. Temporal fusion transformers for interpretable multi-horizon time series forecasting, 2020.","title":"Temporal fusion transformers for interpretable multi-horizon time series forecasting","authors":"Bryan Lim, Sercan O Arik, Nicolas Loeff, Tomas Pfister","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Mukherjee et al., 2018] Srayanta Mukherjee, Devashish Shankar, Atin Ghosh, Nilam Tathawadekar, Pramod Kom- palli, Sunita Sarawagi, and Krishnendu Chaudhury. AR- MDN: associative and recurrent mixture density networks for eretail demand forecasting. CoRR, abs/1803.03800, 2018.","title":"AR-MDN: associative and recurrent mixture density networks for eretail demand forecasting","authors":"Srayanta Mukherjee, Devashish Shankar, Atin Ghosh, Nilam Tathawadekar, Pramod Kompalli, Sunita Sarawagi, Krishnendu Chaudhury","year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Oliva et al., 2017] Junier B. Oliva, Barnabás Póczos, and Jeff G. Schneider. The statistical recurrent unit. In ICML, pages 2671-2680, 2017.","title":"The statistical recurrent unit","authors":"B Junier, Barnabás Oliva, Jeff G Póczos, Schneider","year":"2017","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Oreshkin et al., 2019] Boris N Oreshkin, Dmitri Carpov, Nicolas Chapados, and Yoshua Bengio. N-beats: Neural basis expansion analysis for interpretable time series fore- casting. arXiv preprint arXiv:1905.10437, 2019.","title":"N-beats: Neural basis expansion analysis for interpretable time series forecasting","authors":"Dmitri Boris N Oreshkin, Nicolas Carpov, Yoshua Chapados, Bengio","year":"2019","venue":null,"doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Rangapuram et al., 2018] Syama Sundar Rangapuram, Matthias W Seeger, Jan Gasthaus, Lorenzo Stella, Yuyang Wang, and Tim Januschowski. Deep state space models for time series forecasting. In S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, and R. Gar- nett, editors, Advances in Neural Information Processing Systems 31, pages 7796-7805. 2018.","title":"Deep state space models for time series forecasting","authors":"Syama Sundar Rangapuram, Matthias W Seeger, Jan Gasthaus, Lorenzo Stella, Yuyang Wang, Tim Januschowski","year":"2018","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Rangapuram et al., 2021] Syama Sundar Rangapuram, Lu- cien D Werner, Konstantinos Benidis, Pedro Mercado, Jan Gasthaus, and Tim Januschowski. End-to-end learning of coherent probabilistic forecasts for hierarchical time se- ries. In International Conference on Machine Learning, pages 8832-8843. PMLR, 2021.","title":"End-to-end learning of coherent probabilistic forecasts for hierarchical time series","authors":"Syama Sundar Rangapuram, Lucien D Werner, Konstantinos Benidis, Pedro Mercado, Jan Gasthaus, Tim Januschowski","year":"2021","venue":"International Conference on Machine Learning","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Salinas et al., 2019] David Salinas, Michael Bohlke- Schneider, Laurent Callot, Roberto Medico, and Jan Gasthaus. High-dimensional multivariate forecasting with low-rank gaussian copula processes. In Advances in Neu- ral Information Processing Systems, pages 6827-6837, 2019.","title":"High-dimensional multivariate forecasting with low-rank gaussian copula processes","authors":"David Salinas, Michael Bohlke-Schneider, Laurent Callot, Roberto Medico, Jan Gasthaus","year":"2019","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Taieb et al., 2017] Souhaib Ben Taieb, James W Taylor, and Rob J Hyndman. Coherent probabilistic forecasts for hier- archical time series. In International Conference on Ma- chine Learning, pages 3348-3357, 2017.","title":"Coherent probabilistic forecasts for hierarchical time series","authors":"Taieb","year":"2017","venue":"International Conference on Machine Learning","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Wen et al., 2017] Ruofeng Wen, Kari Torkkola, and Balakr- ishnan Narayanaswamy. A multi-horizon quantile recur- rent forecaster. arXiv preprint arXiv:1711.11053, 2017.","title":"Ruofeng Wen, Kari Torkkola, and Balakrishnan Narayanaswamy. A multi-horizon quantile recurrent forecaster","authors":"Wen","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Wickramasuriya et al., 2019] Shanika L. Wickramasuriya, George Athanasopoulos, and Rob J. Hyndman. Optimal forecast reconciliation for hierarchical and grouped time series through trace minimization. Journal of the Ameri- can Statistical Association, 114(526):804-819, 2019.","title":"Optimal forecast reconciliation for hierarchical and grouped time series through trace minimization","authors":"Shanika L Wickramasuriya, George Athanasopoulos, Rob J Hyndman","year":"2019","venue":"Journal of the American Statistical Association","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Zhou et al., 2020] Haoyi Zhou, Shanghang Zhang, Jieqi Peng, Shuai Zhang, Jianxin Li, Hui Xiong, and Wan- cai Zhang. Informer: Beyond efficient transformer for long sequence time-series forecasting. arXiv preprint arXiv:2012.07436, 2020.","title":"Informer: Beyond efficient transformer for long sequence time-series forecasting","authors":"Zhou","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}