On the Impact of Random Seeds on the Fairness of Clinical Classifiers
| Autor: | Silvio Amir, Byron C. Wallace, Jan-Willem van de Meent |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
0303 health sciences
Computer Science - Computation and Language Computer science Yield (finance) Random seed Small sample 010501 environmental sciences Health records 01 natural sciences 03 medical and health sciences Large networks Statistics Overall performance Limit (mathematics) 030304 developmental biology 0105 earth and related environmental sciences |
| Zdroj: | NAACL-HLT |
| Popis: | Recent work has shown that fine-tuning large networks is surprisingly sensitive to changes in random seed(s). We explore the implications of this phenomenon for model fairness across demographic groups in clinical prediction tasks over electronic health records (EHR) in MIMIC-III -- the standard dataset in clinical NLP research. Apparent subgroup performance varies substantially for seeds that yield similar overall performance, although there is no evidence of a trade-off between overall and subgroup performance. However, we also find that the small sample sizes inherent to looking at intersections of minority groups and somewhat rare conditions limit our ability to accurately estimate disparities. Further, we find that jointly optimizing for high overall performance and low disparities does not yield statistically significant improvements. Our results suggest that fairness work using MIMIC-III should carefully account for variations in apparent differences that may arise from stochasticity and small sample sizes. Comment: Accepted for publication at NAACL 2021 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|