MultiMix TFT: A Multi-task Mixed-Frequency Framework with Temporal Fusion Transformers

Boje Deforce, Bart Baesens, Jan Diels, Estefanía Serral Asensio
Proceedings of The 2nd Conference on Lifelong Learning Agents, PMLR 232:586-600, 2023.

Abstract

Multi-task learning (MTL) has been increasingly recognized as an effective paradigm in time-series analysis for forecasting multiple related tasks concurrently. Prior MTL frameworks for time-series forecasting have typically been devised for tasks that share the same regular time frequencies. However, numerous real-world scenarios entail tasks measured at mixed, and often irregular, time frequencies. We propose a multi-task mixed-frequency (MultiMix) learning framework for time-series forecasting that addresses the challenges of mixed-frequency scenarios where tasks are measured at different and/or irregular time intervals. Our proposed framework leverages the relationships between mixed-frequency tasks to improve accuracy and robustness of time-series forecasting across tasks. The MultiMix framework is implemented using the state-of-the-art Temporal Fusion Transformer (TFT) and is evaluated in smart irrigation, where predicting mid-day stem water potential and soil water potential pose critical challenges. The MultiMix TFT enables joint forecasting of stem water potential, measured sparsely on irregular and infrequent time intervals, and soil water potential, measured on a daily time interval. The results show substantial improvements in stem water potential prediction over state-of-the-art baselines while achieving comparable performance for soil water potential. These results confirm the effectiveness of the proposed framework for addressing the mixed-frequency time-series forecasting problem in real-world settings. Code will be made available upon publication.

Cite this Paper

BibTeX
@InProceedings{pmlr-v232-deforce23a, title = {MultiMix TFT: A Multi-task Mixed-Frequency Framework with Temporal Fusion Transformers}, author = {Deforce, Boje and Baesens, Bart and Diels, Jan and Asensio, Estefan\'ia Serral}, booktitle = {Proceedings of The 2nd Conference on Lifelong Learning Agents}, pages = {586--600}, year = {2023}, editor = {Chandar, Sarath and Pascanu, Razvan and Sedghi, Hanie and Precup, Doina}, volume = {232}, series = {Proceedings of Machine Learning Research}, month = {22--25 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v232/deforce23a/deforce23a.pdf}, url = {https://proceedings.mlr.press/v232/deforce23a.html}, abstract = {Multi-task learning (MTL) has been increasingly recognized as an effective paradigm in time-series analysis for forecasting multiple related tasks concurrently. Prior MTL frameworks for time-series forecasting have typically been devised for tasks that share the same regular time frequencies. However, numerous real-world scenarios entail tasks measured at mixed, and often irregular, time frequencies. We propose a multi-task mixed-frequency (MultiMix) learning framework for time-series forecasting that addresses the challenges of mixed-frequency scenarios where tasks are measured at different and/or irregular time intervals. Our proposed framework leverages the relationships between mixed-frequency tasks to improve accuracy and robustness of time-series forecasting across tasks. The MultiMix framework is implemented using the state-of-the-art Temporal Fusion Transformer (TFT) and is evaluated in smart irrigation, where predicting mid-day stem water potential and soil water potential pose critical challenges. The MultiMix TFT enables joint forecasting of stem water potential, measured sparsely on irregular and infrequent time intervals, and soil water potential, measured on a daily time interval. The results show substantial improvements in stem water potential prediction over state-of-the-art baselines while achieving comparable performance for soil water potential. These results confirm the effectiveness of the proposed framework for addressing the mixed-frequency time-series forecasting problem in real-world settings. Code will be made available upon publication.} }
Endnote
%0 Conference Paper %T MultiMix TFT: A Multi-task Mixed-Frequency Framework with Temporal Fusion Transformers %A Boje Deforce %A Bart Baesens %A Jan Diels %A Estefanía Serral Asensio %B Proceedings of The 2nd Conference on Lifelong Learning Agents %C Proceedings of Machine Learning Research %D 2023 %E Sarath Chandar %E Razvan Pascanu %E Hanie Sedghi %E Doina Precup %F pmlr-v232-deforce23a %I PMLR %P 586--600 %U https://proceedings.mlr.press/v232/deforce23a.html %V 232 %X Multi-task learning (MTL) has been increasingly recognized as an effective paradigm in time-series analysis for forecasting multiple related tasks concurrently. Prior MTL frameworks for time-series forecasting have typically been devised for tasks that share the same regular time frequencies. However, numerous real-world scenarios entail tasks measured at mixed, and often irregular, time frequencies. We propose a multi-task mixed-frequency (MultiMix) learning framework for time-series forecasting that addresses the challenges of mixed-frequency scenarios where tasks are measured at different and/or irregular time intervals. Our proposed framework leverages the relationships between mixed-frequency tasks to improve accuracy and robustness of time-series forecasting across tasks. The MultiMix framework is implemented using the state-of-the-art Temporal Fusion Transformer (TFT) and is evaluated in smart irrigation, where predicting mid-day stem water potential and soil water potential pose critical challenges. The MultiMix TFT enables joint forecasting of stem water potential, measured sparsely on irregular and infrequent time intervals, and soil water potential, measured on a daily time interval. The results show substantial improvements in stem water potential prediction over state-of-the-art baselines while achieving comparable performance for soil water potential. These results confirm the effectiveness of the proposed framework for addressing the mixed-frequency time-series forecasting problem in real-world settings. Code will be made available upon publication.
APA
Deforce, B., Baesens, B., Diels, J. & Asensio, E.S.. (2023). MultiMix TFT: A Multi-task Mixed-Frequency Framework with Temporal Fusion Transformers. Proceedings of The 2nd Conference on Lifelong Learning Agents, in Proceedings of Machine Learning Research 232:586-600 Available from https://proceedings.mlr.press/v232/deforce23a.html.

Related Material