Taming the Recent-Data Bias: Towards Robust Time Series Forecasting with Global Context

Longlong Xu, Zeyan Li, Xiao He, Zhaoyang Yu, Changhua Pei, Zhe Xie, Zijun Dou, Tieying Zhang, Dan Pei
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:141265-141295, 2026.

Abstract

Time series forecasting plays a vital role in numerous domains. However, real-world time series are frequently contaminated by noise, missing values, and anomalies, posing significant challenges to reliable forecasting. In this work, we first systematically investigate a fundamental limitation prevalent in existing forecasting methods: an excessive reliance on the most recent observations—termed "recent-data bias". This bias renders forecasts highly vulnerable to perturbations in recent data, severely undermining prediction reliability. To address this issue, we propose TameR, a novel approach for robust time series forecasting that effectively mitigates recent-data bias via enhancing the utilization of global context. Specifically, it employs a basis-aligned randomized sampling strategy to reduce dependence on any specific recent data. Furthermore, TameR incorporates a learnable periodicity extraction module coupled with a two-stage learning protocol to robustly separate periodic patterns from the sampled residual components. Comprehensive experiments demonstrate that TameR significantly outperforms state-of-the-art methods in robustness against diverse perturbation scenarios, while achieving comparable accuracy on clean data. Code is available at https://github.com/NetManAIOps/TameR.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-xu26ab, title = {Taming the Recent-Data Bias: Towards Robust Time Series Forecasting with Global Context}, author = {Xu, Longlong and Li, Zeyan and He, Xiao and Yu, Zhaoyang and Pei, Changhua and Xie, Zhe and Dou, Zijun and Zhang, Tieying and Pei, Dan}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {141265--141295}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/xu26ab/xu26ab.pdf}, url = {https://proceedings.mlr.press/v306/xu26ab.html}, abstract = {Time series forecasting plays a vital role in numerous domains. However, real-world time series are frequently contaminated by noise, missing values, and anomalies, posing significant challenges to reliable forecasting. In this work, we first systematically investigate a fundamental limitation prevalent in existing forecasting methods: an excessive reliance on the most recent observations—termed "recent-data bias". This bias renders forecasts highly vulnerable to perturbations in recent data, severely undermining prediction reliability. To address this issue, we propose TameR, a novel approach for robust time series forecasting that effectively mitigates recent-data bias via enhancing the utilization of global context. Specifically, it employs a basis-aligned randomized sampling strategy to reduce dependence on any specific recent data. Furthermore, TameR incorporates a learnable periodicity extraction module coupled with a two-stage learning protocol to robustly separate periodic patterns from the sampled residual components. Comprehensive experiments demonstrate that TameR significantly outperforms state-of-the-art methods in robustness against diverse perturbation scenarios, while achieving comparable accuracy on clean data. Code is available at https://github.com/NetManAIOps/TameR.} }
Endnote
%0 Conference Paper %T Taming the Recent-Data Bias: Towards Robust Time Series Forecasting with Global Context %A Longlong Xu %A Zeyan Li %A Xiao He %A Zhaoyang Yu %A Changhua Pei %A Zhe Xie %A Zijun Dou %A Tieying Zhang %A Dan Pei %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-xu26ab %I PMLR %P 141265--141295 %U https://proceedings.mlr.press/v306/xu26ab.html %V 306 %X Time series forecasting plays a vital role in numerous domains. However, real-world time series are frequently contaminated by noise, missing values, and anomalies, posing significant challenges to reliable forecasting. In this work, we first systematically investigate a fundamental limitation prevalent in existing forecasting methods: an excessive reliance on the most recent observations—termed "recent-data bias". This bias renders forecasts highly vulnerable to perturbations in recent data, severely undermining prediction reliability. To address this issue, we propose TameR, a novel approach for robust time series forecasting that effectively mitigates recent-data bias via enhancing the utilization of global context. Specifically, it employs a basis-aligned randomized sampling strategy to reduce dependence on any specific recent data. Furthermore, TameR incorporates a learnable periodicity extraction module coupled with a two-stage learning protocol to robustly separate periodic patterns from the sampled residual components. Comprehensive experiments demonstrate that TameR significantly outperforms state-of-the-art methods in robustness against diverse perturbation scenarios, while achieving comparable accuracy on clean data. Code is available at https://github.com/NetManAIOps/TameR.
APA
Xu, L., Li, Z., He, X., Yu, Z., Pei, C., Xie, Z., Dou, Z., Zhang, T. & Pei, D.. (2026). Taming the Recent-Data Bias: Towards Robust Time Series Forecasting with Global Context. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:141265-141295 Available from https://proceedings.mlr.press/v306/xu26ab.html.

Related Material