Adaptive Sampling for Probabilistic Forecasting Under Distribution Shift
Abstract
The world is not static: This causes real-world time series to change over time through external, and potentially disruptive, events such as macroeconomic cycles or the COVID-19 pandemic. We present an adaptive sampling strategy that selects the part of the time series history that is relevant for forecasting. We achieve this by learning a discrete distribution over relevant time steps by Bayesian optimization. We instantiate this idea with a two-step method that is pre-trained with uniform sampling and then training a lightweight adaptive architecture with adaptive sam- pling. We show with synthetic and real-world experiments that this method adapts to distribution shift and significantly reduces the forecasting error of the base model for three out of five datasets.
Cite
Text
Masserano et al. "Adaptive Sampling for Probabilistic Forecasting Under Distribution Shift." NeurIPS 2022 Workshops: DistShift, 2022.Markdown
[Masserano et al. "Adaptive Sampling for Probabilistic Forecasting Under Distribution Shift." NeurIPS 2022 Workshops: DistShift, 2022.](https://mlanthology.org/neuripsw/2022/masserano2022neuripsw-adaptive/)BibTeX
@inproceedings{masserano2022neuripsw-adaptive,
title = {{Adaptive Sampling for Probabilistic Forecasting Under Distribution Shift}},
author = {Masserano, Luca and Rangapuram, Syama Sundar and Kapoor, Shubham and Nirwan, Rajbir Singh and Park, Youngsuk and Bohlke-Schneider, Michael},
booktitle = {NeurIPS 2022 Workshops: DistShift},
year = {2022},
url = {https://mlanthology.org/neuripsw/2022/masserano2022neuripsw-adaptive/}
}