Temporal Label Smoothing for Early Prediction of Adverse Events

Hugo Yèche, Alizée Pace, Gunnar Ratsch, Rita Kuznetsova

Keywords: Healthcare, Time-Series, Label Smoothing, Deep Learning, Application

TL;DR: Modulating label smoothing strength over time to reflect signal noise patterns and clinical priorities significantly improves deep learning model performance in the prediction of adverse medical events.

Abstract: Models that can predict adverse events ahead of time with low false-alarm rates are critical to the acceptance of decision support systems in the medical community. This challenging machine learning task remains typically treated as a simple binary classification, with few bespoke methods proposed to leverage temporal dependency across samples. We propose Temporal Label Smoothing (TLS), a novel learning strategy that modulates smoothing strength as a function of proximity to the event of interest. This regularization technique reduces model confidence at the class boundary, where the signal is often noisy or uninformative, thus allowing training to focus on clinically informative data points away from this boundary region. From a theoretical perspective, we also show that our method can be framed as an extension of multi-horizon prediction, a learning heuristic proposed in other early prediction work. TLS empirically matches or outperforms all competitor methods across all evaluation measures on various early prediction benchmark tasks. In particular, our approach significantly improves performance on clinically-relevant metrics such as event recall under low false-alarm rates.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Machine Learning for Sciences (eg biology, physics, health sciences, social sciences, climate/sustainability )

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2208.13764/code)