Theory and Algorithm for Batch Distribution Drift Problems
Abstract
We study a problem of gradual \emph{batch distribution drift} motivated by several applications, which consists of determining an accurate predictor for a target time segment, for which a moderate amount of labeled samples are at one's disposal, while leveraging past segments for which substantially more labeled samples are available. We give new algorithms for this problem guided by a new theoretical analysis and generalization bounds derived for this scenario. Additionally, we report the results of extensive experiments demonstrating the benefits of our drifting algorithm, including comparisons with natural baselines.
Cite
Text
Awasthi et al. "Theory and Algorithm for Batch Distribution Drift Problems." NeurIPS 2022 Workshops: DistShift, 2022.Markdown
[Awasthi et al. "Theory and Algorithm for Batch Distribution Drift Problems." NeurIPS 2022 Workshops: DistShift, 2022.](https://mlanthology.org/neuripsw/2022/awasthi2022neuripsw-theory/)BibTeX
@inproceedings{awasthi2022neuripsw-theory,
title = {{Theory and Algorithm for Batch Distribution Drift Problems}},
author = {Awasthi, Pranjal and Cortes, Corinna and Mohri, Christopher},
booktitle = {NeurIPS 2022 Workshops: DistShift},
year = {2022},
url = {https://mlanthology.org/neuripsw/2022/awasthi2022neuripsw-theory/}
}