{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T14:04:00Z","timestamp":1774620240866,"version":"3.50.1"},"reference-count":57,"publisher":"Association for Computing Machinery (ACM)","issue":"7","content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["Proc. VLDB Endow."],"published-print":{"date-parts":[[2024,3]]},"abstract":"<jats:p>\n            Time series data, including univariate and multivariate ones, are characterized by unique composition and complex multi-scale temporal variations. They often require special consideration of decomposition and multi-scale modeling to analyze. Existing deep learning methods on this best fit to univariate time series only, and have not sufficiently considered sub-series modeling and decomposition completeness. To address these challenges, we propose MSD-Mixer, a\n            <jats:bold>M<\/jats:bold>\n            ulti-\n            <jats:bold>S<\/jats:bold>\n            cale\n            <jats:bold>D<\/jats:bold>\n            ecomposition MLP-\n            <jats:bold>Mixer<\/jats:bold>\n            , which learns to explicitly decompose and represent the input time series in its different layers. To handle the multi-scale temporal patterns and multivariate dependencies, we propose a novel temporal patching approach to model the time series as multi-scale patches, and employ MLPs to capture intra- and inter-patch variations and channel-wise correlations. In addition, we propose a novel loss function to constrain both the mean and the autocorrelation of the decomposition residual for better decomposition completeness. Through extensive experiments on various real-world datasets for five common time series analysis tasks, we demonstrate that MSD-Mixer consistently and significantly outperforms other state-of-the-art algorithms with better efficiency.\n          <\/jats:p>","DOI":"10.14778\/3654621.3654637","type":"journal-article","created":{"date-parts":[[2024,5,30]],"date-time":"2024-05-30T22:21:08Z","timestamp":1717107668000},"page":"1723-1736","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":37,"title":["A Multi-Scale Decomposition MLP-Mixer for Time Series Analysis"],"prefix":"10.14778","volume":"17","author":[{"given":"Shuhan","family":"Zhong","sequence":"first","affiliation":[{"name":"Department of Computer Science and Engineering, The Hong Kong University of Science and Technology"}]},{"given":"Sizhe","family":"Song","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, The Hong Kong University of Science and Technology"}]},{"given":"Weipeng","family":"Zhuo","sequence":"additional","affiliation":[{"name":"Guangdong Provincial Key Laboratory IRADS and Department of Computer Science, BNU-HKBU United International College"}]},{"given":"Guanyao","family":"Li","sequence":"additional","affiliation":[{"name":"Guangdong Enterprise Key Laboratory for Urban Sensing, Monitoring and Early Warning, Guangzhou Urban Planning and Design Survey Research Institute"}]},{"given":"Yang","family":"Liu","sequence":"additional","affiliation":[{"name":"Guangdong Enterprise Key Laboratory for Urban Sensing, Monitoring and Early Warning, Guangzhou Urban Planning and Design Survey Research Institute"}]},{"given":"S.-H. Gary","family":"Chan","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, The Hong Kong University of Science and Technology"}]}],"member":"320","published-online":{"date-parts":[[2024,5,30]]},"reference":[{"key":"e_1_2_1_1_1","volume-title":"Jason Lines, Michael Flynn, James Large, Aaron Bostrom, Paul Southam, and Eamonn Keogh.","author":"Bagnall Anthony","year":"2018","unstructured":"Anthony Bagnall, Hoang Anh Dau, Jason Lines, Michael Flynn, James Large, Aaron Bostrom, Paul Southam, and Eamonn Keogh. 2018. The UEA multivariate time series classification archive, 2018. arXiv:1811.00075 [cs.LG]"},{"key":"e_1_2_1_2_1","unstructured":"Shaojie Bai J. Zico Kolter and Vladlen Koltun. 2018. An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling. arXiv:1803.01271 [cs.LG]"},{"key":"e_1_2_1_3_1","volume-title":"MSTL: A Seasonal-Trend Decomposition Algorithm for Time Series with Multiple Seasonal Patterns. arXiv:2107.13462 [stat.AP]","author":"Bandara Kasun","year":"2021","unstructured":"Kasun Bandara, Rob J Hyndman, and Christoph Bergmeir. 2021. MSTL: A Seasonal-Trend Decomposition Algorithm for Time Series with Multiple Seasonal Patterns. arXiv:2107.13462 [stat.AP]"},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.14778\/3476249.3476300"},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.14778\/3137765.3137775"},{"key":"e_1_2_1_6_1","volume-title":"Max Mergenthaler Canseco, and Artur Dubrawski.","author":"Challu Cristian","year":"2023","unstructured":"Cristian Challu, Kin G Olivares, Boris N Oreshkin, Federico Garza Ramirez, Max Mergenthaler Canseco, and Artur Dubrawski. 2023. NHITS: Neural Hierarchical Interpolation for Time Series Forecasting., 6989--6997 pages."},{"key":"e_1_2_1_7_1","doi-asserted-by":"crossref","unstructured":"Mingyue Cheng Qi Liu Zhiding Liu Zhi Li Yucong Luo and Enhong Chen. 2023. FormerTime: Hierarchical Multi-Scale Representations for Multivariate Time Series Classification. arXiv:2302.09818 [cs.LG]","DOI":"10.1145\/3543507.3583205"},{"key":"e_1_2_1_8_1","volume-title":"Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (Washington DC, USA) (KDD '22)","author":"Chowdhury Ranak Roy","year":"2022","unstructured":"Ranak Roy Chowdhury, Xiyuan Zhang, Jingbo Shang, Rajesh K. Gupta, and Dezhi Hong. 2022. TARNet: Task-Aware Reconstruction for Time-Series Transformer. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (Washington DC, USA) (KDD '22). Association for Computing Machinery, New York, NY, USA, 212--220."},{"key":"e_1_2_1_9_1","first-page":"3","article-title":"STL: A seasonal-trend decomposition","volume":"6","author":"Cleveland Robert B","year":"1990","unstructured":"Robert B Cleveland, William S Cleveland, Jean E McRae, and Irma Terpenning. 1990. STL: A seasonal-trend decomposition. J. Off. Stat 6, 1 (1990), 3--73.","journal-title":"J. Off. Stat"},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.14778\/3489496.3489503"},{"key":"e_1_2_1_11_1","volume-title":"Seasonal adjustment methods and real time trend-cycle estimation","author":"Dagum Estela Bee","unstructured":"Estela Bee Dagum and Silvia Bianconcini. 2016. Seasonal adjustment methods and real time trend-cycle estimation. Springer, New York, NY, USA."},{"key":"e_1_2_1_12_1","volume-title":"Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining","author":"Dempster Angus","unstructured":"Angus Dempster, Daniel F. Schmidt, and Geoffrey I. Webb. 2021. MiniRocket: A Very Fast (Almost) Deterministic Transform for Time Series Classification. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining (Virtual Event, Singapore) (KDD '21). Association for Computing Machinery, New York, NY, USA, 248--257."},{"key":"e_1_2_1_13_1","volume-title":"STR: A seasonal-trend decomposition procedure based on regression. Monash econometrics and business statistics working papers 13, 15","author":"Dokumentov Alexander","year":"2015","unstructured":"Alexander Dokumentov, Rob J Hyndman, et al. 2015. STR: A seasonal-trend decomposition procedure based on regression. Monash econometrics and business statistics working papers 13, 15 (2015), 2015--13."},{"key":"e_1_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.14778\/3229863.3229878"},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.14778\/3457390.3457394"},{"key":"e_1_2_1_16_1","doi-asserted-by":"crossref","unstructured":"Lu Han Han-Jia Ye and De-Chuan Zhan. 2023. The Capacity and Robustness Trade-off: Revisiting the Channel Independent Strategy for Multivariate Time Series Forecasting. arXiv:2304.05206 [cs.LG]","DOI":"10.1109\/TKDE.2024.3400008"},{"key":"e_1_2_1_17_1","volume-title":"International journal of forecasting 20, 1","author":"Holt Charles C","year":"2004","unstructured":"Charles C Holt. 2004. Forecasting seasonals and trends by exponentially weighted moving averages. International journal of forecasting 20, 1 (2004), 5--10."},{"key":"e_1_2_1_18_1","volume-title":"Forecasting: principles and practice. OTexts","author":"Hyndman Rob J","unstructured":"Rob J Hyndman and George Athanasopoulos. 2018. Forecasting: principles and practice. OTexts, Melbourne, Australia."},{"key":"e_1_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10618-020-00710-y"},{"key":"e_1_2_1_20_1","volume-title":"Multivariate LSTM-FCNs for time series classification. Neural networks 116","author":"Karim Fazle","year":"2019","unstructured":"Fazle Karim, Somshubra Majumdar, Houshang Darabi, and Samuel Harford. 2019. Multivariate LSTM-FCNs for time series classification. Neural networks 116 (2019), 237--245."},{"key":"e_1_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.14778\/3377369.3377383"},{"key":"e_1_2_1_22_1","volume-title":"Reformer: The Efficient Transformer. arXiv:2001.04451 [cs.LG]","author":"Kitaev Nikita","year":"2020","unstructured":"Nikita Kitaev, \u0141ukasz Kaiser, and Anselm Levskaya. 2020. Reformer: The Efficient Transformer. arXiv:2001.04451 [cs.LG]"},{"key":"e_1_2_1_23_1","unstructured":"Gustav Larsson Michael Maire and Gregory Shakhnarovich. 2017. FractalNet: Ultra-Deep Neural Networks without Residuals. arXiv:1605.07648 [cs.CV]"},{"key":"e_1_2_1_24_1","volume-title":"Austin Reiter, and Gregory D. Hager","author":"Lea Colin","year":"2016","unstructured":"Colin Lea, Michael D. Flynn, Rene Vidal, Austin Reiter, and Gregory D. Hager. 2016. Temporal Convolutional Networks for Action Segmentation and Detection. arXiv:1611.05267 [cs.CV]"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.14778\/1453856.1453972"},{"key":"e_1_2_1_26_1","volume-title":"Modeling temporal patterns with dilated convolutions for time-series forecasting. ACM Transactions on Knowledge Discovery from Data (TKDD) 16, 1","author":"Li Yangfan","year":"2021","unstructured":"Yangfan Li, Kenli Li, Cen Chen, Xu Zhou, Zeng Zeng, and Keqin Li. 2021. Modeling temporal patterns with dilated convolutions for time-series forecasting. ACM Transactions on Knowledge Discovery from Data (TKDD) 16, 1 (2021), 1--22."},{"key":"e_1_2_1_27_1","first-page":"5816","article-title":"Scinet: Time series modeling and forecasting with sample convolution and interaction","volume":"35","author":"Liu Minhao","year":"2022","unstructured":"Minhao Liu, Ailing Zeng, Muxi Chen, Zhijian Xu, Qiuxia Lai, Lingna Ma, and Qiang Xu. 2022. Scinet: Time series modeling and forecasting with sample convolution and interaction. Advances in Neural Information Processing Systems 35 (2022), 5816--5828.","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_1_28_1","volume-title":"Non-stationary Transformers: Exploring the Stationarity in Time Series Forecasting. In Advances in Neural Information Processing Systems","author":"Liu Yong","year":"2022","unstructured":"Yong Liu, Haixu Wu, Jianmin Wang, and Mingsheng Long. 2022. Non-stationary Transformers: Exploring the Stationarity in Time Series Forecasting. In Advances in Neural Information Processing Systems, S. Koyejo, S. Mohamed, A. Agarwal, D. Belgrave, K. Cho, and A. Oh (Eds.), Vol. 35. Curran Associates, Inc., New Orleans, LA, USA, 9881--9893."},{"key":"e_1_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijforecast.2019.04.014"},{"key":"e_1_2_1_30_1","doi-asserted-by":"crossref","unstructured":"Xiaoye Miao Yangyang Wu Jun Wang Yunjun Gao Xudong Mao and Jianwei Yin. 2021. Generative semi-supervised learning for multivariate time series imputation. 8983--8991 pages.","DOI":"10.1609\/aaai.v35i10.17086"},{"key":"e_1_2_1_31_1","unstructured":"Yuqi Nie Nam H. Nguyen Phanwadee Sinthong and Jayant Kalagnanam. 2023. A Time Series is Worth 64 Words: Long-term Forecasting with Transformers. arXiv:2211.14730 [cs.LG]"},{"key":"e_1_2_1_32_1","unstructured":"Boris N. Oreshkin Dmitri Carpov Nicolas Chapados and Yoshua Bengio. 2020. N-BEATS: Neural basis expansion analysis for interpretable time series forecasting."},{"key":"e_1_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.14778\/3551793.3551830"},{"key":"e_1_2_1_34_1","doi-asserted-by":"publisher","DOI":"10.14778\/3529337.3529354"},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.1145\/3292500.3330680"},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.14778\/3538598.3538602"},{"key":"e_1_2_1_37_1","volume-title":"Scale-former: Iterative Multi-scale Refining Transformers for Time Series Forecasting. arXiv:2206.04038 [cs.LG]","author":"Shabani Amin","year":"2023","unstructured":"Amin Shabani, Amir Abdi, Lili Meng, and Tristan Sylvain. 2023. Scale-former: Iterative Multi-scale Refining Transformers for Time Series Forecasting. arXiv:2206.04038 [cs.LG]"},{"key":"e_1_2_1_38_1","volume-title":"Proceedings of the 2015 SIAM International Conference on Data Mining (SDM). SIAM","author":"Shokoohi-Yekta Mohammad","year":"2015","unstructured":"Mohammad Shokoohi-Yekta, Jun Wang, and Eamonn Keogh. 2015. On the Non-Trivial Generalization of Dynamic Time Warping to the Multi-Dimensional Case. In Proceedings of the 2015 SIAM International Conference on Data Mining (SDM). SIAM, Vancouver, BC, Canada, 289--297."},{"key":"e_1_2_1_39_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijforecast.2010.11.002"},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.14778\/3415478.3415518"},{"key":"e_1_2_1_41_1","unstructured":"Ashish Vaswani Noam Shazeer Niki Parmar Jakob Uszkoreit Llion Jones Aidan N. Gomez Lukasz Kaiser and Illia Polosukhin. 2017. Attention Is All You Need. arXiv:1706.03762 [cs.CL]"},{"key":"e_1_2_1_42_1","volume-title":"Micn: Multi-scale local and global context modeling for long-term series forecasting.","author":"Wang Huiqiang","year":"2023","unstructured":"Huiqiang Wang, Jian Peng, Feihu Huang, Jince Wang, Junhui Chen, and Yifei Xiao. 2023. Micn: Multi-scale local and global context modeling for long-term series forecasting."},{"key":"e_1_2_1_43_1","unstructured":"Qingsong Wen Jingkun Gao Xiaomin Song Liang Sun Huan Xu and Shenghuo Zhu. 2018. RobustSTL: A Robust Seasonal-Trend Decomposition Algorithm for Long Time Series. arXiv:1812.01767 [cs.LG]"},{"key":"e_1_2_1_44_1","doi-asserted-by":"publisher","DOI":"10.1145\/3394486.3403271"},{"key":"e_1_2_1_45_1","doi-asserted-by":"publisher","DOI":"10.1093\/biomet\/84.2.489"},{"key":"e_1_2_1_46_1","volume-title":"Forecasting sales by exponentially weighted moving averages. Management science 6, 3","author":"Winters Peter R","year":"1960","unstructured":"Peter R Winters. 1960. Forecasting sales by exponentially weighted moving averages. Management science 6, 3 (1960), 324--342."},{"key":"e_1_2_1_47_1","unstructured":"Gerald Woo Chenghao Liu Doyen Sahoo Akshat Kumar and Steven Hoi. 2022. ETSformer: Exponential Smoothing Transformers for Time-series Forecasting. arXiv:2202.01381"},{"key":"e_1_2_1_48_1","unstructured":"Haixu Wu Tengge Hu Yong Liu Hang Zhou Jianmin Wang and Mingsheng Long. 2023. TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis. arXiv:2210.02186 [cs.LG]"},{"key":"e_1_2_1_49_1","first-page":"22419","article-title":"Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting","volume":"34","author":"Wu Haixu","year":"2021","unstructured":"Haixu Wu, Jiehui Xu, Jianmin Wang, and Mingsheng Long. 2021. Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting. Advances in Neural Information Processing Systems 34 (2021), 22419--22430.","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_1_50_1","volume-title":"Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy. arXiv:2110.02642 [cs.LG]","author":"Xu Jiehui","year":"2022","unstructured":"Jiehui Xu, Haixu Wu, Jianmin Wang, and Mingsheng Long. 2022. Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy. arXiv:2110.02642 [cs.LG]"},{"key":"e_1_2_1_51_1","unstructured":"Ailing Zeng Muxi Chen Lei Zhang and Qiang Xu. 2022. Are Transformers Effective for Time Series Forecasting? arXiv:2205.13504 [cs.AI]"},{"key":"e_1_2_1_52_1","doi-asserted-by":"publisher","DOI":"10.1145\/3447548.3467401"},{"key":"e_1_2_1_53_1","unstructured":"Tianping Zhang Yizhuo Zhang Wei Cao Jiang Bian Xiaohan Yi Shun Zheng and Jian Li. 2022. Less Is More: Fast Multivariate Time Series Forecasting with Light Sampling-oriented MLP Structures. arXiv:2207.01186 [cs.LG]"},{"key":"e_1_2_1_54_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i04.6165"},{"key":"e_1_2_1_55_1","volume-title":"Crossformer: Transformer Utilizing Cross-Dimension Dependency for Multivariate Time Series Forecasting.","author":"Zhang Yunhao","year":"2023","unstructured":"Yunhao Zhang and Junchi Yan. 2023. Crossformer: Transformer Utilizing Cross-Dimension Dependency for Multivariate Time Series Forecasting."},{"key":"e_1_2_1_56_1","volume-title":"Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting. arXiv:2012.07436 [cs.LG]","author":"Zhou Haoyi","year":"2021","unstructured":"Haoyi Zhou, Shanghang Zhang, Jieqi Peng, Shuai Zhang, Jianxin Li, Hui Xiong, and Wancai Zhang. 2021. Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting. arXiv:2012.07436 [cs.LG]"},{"key":"e_1_2_1_57_1","unstructured":"Tian Zhou Ziqing Ma Qingsong Wen Xue Wang Liang Sun and Rong Jin. 2022. FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecasting. arXiv:2201.12740[cs.LG]"}],"container-title":["Proceedings of the VLDB Endowment"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.14778\/3654621.3654637","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,5,30]],"date-time":"2024-05-30T22:26:37Z","timestamp":1717107997000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.14778\/3654621.3654637"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,3]]},"references-count":57,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2024,3]]}},"alternative-id":["10.14778\/3654621.3654637"],"URL":"https:\/\/doi.org\/10.14778\/3654621.3654637","relation":{},"ISSN":["2150-8097"],"issn-type":[{"value":"2150-8097","type":"print"}],"subject":[],"published":{"date-parts":[[2024,3]]},"assertion":[{"value":"2024-05-30","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}