FlashST: A Simple and Universal Prompt-Tuning Framework for Traffic Prediction

Zhonghang Li, Lianghao Xia, Yong Xu, Chao Huang
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:28978-28988, 2024.

Abstract

The objective of traffic prediction is to accurately forecast and analyze the dynamics of transportation patterns, considering both space and time. However, the presence of distribution shift poses a significant challenge in this field, as existing models struggle to generalize well when faced with test data that significantly differs from the training distribution. To tackle this issue, this paper introduces a simple and universal spatio-temporal prompt-tuning framework-FlashST, which adapts pre-trained models to the specific characteristics of diverse downstream datasets, improving generalization in diverse traffic prediction scenarios. Specifically, the FlashST framework employs a lightweight spatio-temporal prompt network for in-context learning, capturing spatio-temporal invariant knowledge and facilitating effective adaptation to diverse scenarios. Additionally, we incorporate a distribution mapping mechanism to align the data distributions of pre-training and downstream data, facilitating effective knowledge transfer in spatio-temporal forecasting. Empirical evaluations demonstrate the effectiveness of our FlashST across different spatio-temporal prediction tasks using diverse urban datasets. Code is available at https://github.com/HKUDS/FlashST.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-li24bw, title = {{F}lash{ST}: A Simple and Universal Prompt-Tuning Framework for Traffic Prediction}, author = {Li, Zhonghang and Xia, Lianghao and Xu, Yong and Huang, Chao}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {28978--28988}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/li24bw/li24bw.pdf}, url = {https://proceedings.mlr.press/v235/li24bw.html}, abstract = {The objective of traffic prediction is to accurately forecast and analyze the dynamics of transportation patterns, considering both space and time. However, the presence of distribution shift poses a significant challenge in this field, as existing models struggle to generalize well when faced with test data that significantly differs from the training distribution. To tackle this issue, this paper introduces a simple and universal spatio-temporal prompt-tuning framework-FlashST, which adapts pre-trained models to the specific characteristics of diverse downstream datasets, improving generalization in diverse traffic prediction scenarios. Specifically, the FlashST framework employs a lightweight spatio-temporal prompt network for in-context learning, capturing spatio-temporal invariant knowledge and facilitating effective adaptation to diverse scenarios. Additionally, we incorporate a distribution mapping mechanism to align the data distributions of pre-training and downstream data, facilitating effective knowledge transfer in spatio-temporal forecasting. Empirical evaluations demonstrate the effectiveness of our FlashST across different spatio-temporal prediction tasks using diverse urban datasets. Code is available at https://github.com/HKUDS/FlashST.} }
Endnote
%0 Conference Paper %T FlashST: A Simple and Universal Prompt-Tuning Framework for Traffic Prediction %A Zhonghang Li %A Lianghao Xia %A Yong Xu %A Chao Huang %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-li24bw %I PMLR %P 28978--28988 %U https://proceedings.mlr.press/v235/li24bw.html %V 235 %X The objective of traffic prediction is to accurately forecast and analyze the dynamics of transportation patterns, considering both space and time. However, the presence of distribution shift poses a significant challenge in this field, as existing models struggle to generalize well when faced with test data that significantly differs from the training distribution. To tackle this issue, this paper introduces a simple and universal spatio-temporal prompt-tuning framework-FlashST, which adapts pre-trained models to the specific characteristics of diverse downstream datasets, improving generalization in diverse traffic prediction scenarios. Specifically, the FlashST framework employs a lightweight spatio-temporal prompt network for in-context learning, capturing spatio-temporal invariant knowledge and facilitating effective adaptation to diverse scenarios. Additionally, we incorporate a distribution mapping mechanism to align the data distributions of pre-training and downstream data, facilitating effective knowledge transfer in spatio-temporal forecasting. Empirical evaluations demonstrate the effectiveness of our FlashST across different spatio-temporal prediction tasks using diverse urban datasets. Code is available at https://github.com/HKUDS/FlashST.
APA
Li, Z., Xia, L., Xu, Y. & Huang, C.. (2024). FlashST: A Simple and Universal Prompt-Tuning Framework for Traffic Prediction. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:28978-28988 Available from https://proceedings.mlr.press/v235/li24bw.html.

Related Material