{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/how-features-benefit-parallel-series","title":"How Features Benefit: Parallel Series Embedding for Multivariate Time Series Forecasting with Transformer","arxiv_id":null,"date":"2022-10-31","proceeding":"ICTAI 2022 10","authors":["Xuande Feng","Zonglin Lyu"],"abstract":"Forecasting time series is an engaging and vital\r\nmathematical topic. Theories and applications in related fields\r\nhave been studied for decades, and deep learning has provided\r\nreliable tools in recent years. Transformer, capable to capture\r\nlonger sequence dependencies, was exploited as a powerful architecture in time series forecasting. While existing work majorly\r\ncontributed to breaking memory bottleneck of Trasnformer, how\r\nto effectively leverage multivariate time series remains barely\r\nfocused. In this work, a novel architecture utilizing a primary\r\nTransformer is proposed to conduct multivariate time series\r\npredictions. Our proposed architecture has two main advantages. Firstly, it accurately predicts multivariate time series with\r\nshorter or longer sequence lengths and steps. We benchmark\r\nour proposed model with various baseline architectures on realworld datasets, and our model improved their performances\r\nsignificantly. Secondly, it can easily be leveraged in Transformerbased variants,","url_abs":"https://ieeexplore.ieee.org/abstract/document/10098079","url_pdf":"https://ieeexplore.ieee.org/abstract/document/10098079","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"how-features-benefit-parallel-series","repo_url":"https://github.com/ZonglinL/ParallelSeries","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"multivariate-time-series-forecasting","task_name":"Multivariate Time Series Forecasting"},{"task_slug":"time-series-1","task_name":"Time Series"},{"task_slug":"time-series-forecasting","task_name":"Time Series Forecasting"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/time-series-forecasting-on-etth1-720-2","task":"Time Series Forecasting","dataset":"ETTh1 (720) Univariate","model":"Parallel Series Transformer","rank_in_archive_order":8,"of":12,"metrics":{"MAE":"0.286","MSE":"0.129"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}