Temporal Label Smoothing for Early Event Prediction


METADATA ONLY
Loading...

Date

2023

Publication Type

Conference Paper

ETH Bibliography

yes

Citations

Altmetric
METADATA ONLY

Data

Rights / License

Abstract

Models that can predict the occurrence of events ahead of time with low false-alarm rates are critical to the acceptance of decision support systems in the medical community. This challenging task is typically treated as a simple binary classification, ignoring temporal dependencies between samples, whereas we propose to exploit this structure. We first introduce a common theoretical framework unifying dynamic survival analysis and early event prediction. Following an analysis of objectives from both fields, we propose Temporal Label Smoothing (TLS), a simpler, yet best-performing method that preserves prediction monotonicity over time. By focusing the objective on areas with a stronger predictive signal, TLS improves performance over all baselines on two large-scale benchmark tasks. Gains are particularly notable along clinically relevant measures, such as event recall at low false-alarm rates. TLS reduces the number of missed events by up to a factor of two over previously used approaches in early event prediction.

Publication status

published

Book title

Proceedings of the 40th International Conference on Machine Learning

Volume

202

Pages / Article No.

39913 - 39938

Publisher

PMLR

Event

40th International Conference on Machine Learning (ICML 2023)

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

Organisational unit

09568 - Rätsch, Gunnar / Rätsch, Gunnar check_circle
09568 - Rätsch, Gunnar / Rätsch, Gunnar check_circle
09664 - Schölkopf, Bernhard / Schölkopf, Bernhard check_circle
02219 - ETH AI Center / ETH AI Center

Notes

Funding

Related publications and datasets

Is new version of: