Accounting for Underspecification in Statistical Claims of Model Superiority
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918186018930688 |
|---|---|
| author | Sanchez, Thomas Gordaliza, Pedro M. Cuadra, Meritxell Bach |
| author_facet | Sanchez, Thomas Gordaliza, Pedro M. Cuadra, Meritxell Bach |
| contents | Machine learning methods are increasingly applied in medical imaging, yet many reported improvements lack statistical robustness: recent works have highlighted that small but significant performance gains are highly likely to be false positives. However, these analyses do not take \emph{underspecification} into account -- the fact that models achieving similar validation scores may behave differently on unseen data due to random initialization or training dynamics. Here, we extend a recent statistical framework modeling false outperformance claims to include underspecification as an additional variance component. Our simulations demonstrate that even modest seed variability ($\sim1\%$) substantially increases the evidence required to support superiority claims. Our findings underscore the need for explicit modeling of training variance when validating medical imaging systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_02453 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Accounting for Underspecification in Statistical Claims of Model Superiority Sanchez, Thomas Gordaliza, Pedro M. Cuadra, Meritxell Bach Machine Learning Image and Video Processing Machine learning methods are increasingly applied in medical imaging, yet many reported improvements lack statistical robustness: recent works have highlighted that small but significant performance gains are highly likely to be false positives. However, these analyses do not take \emph{underspecification} into account -- the fact that models achieving similar validation scores may behave differently on unseen data due to random initialization or training dynamics. Here, we extend a recent statistical framework modeling false outperformance claims to include underspecification as an additional variance component. Our simulations demonstrate that even modest seed variability ($\sim1\%$) substantially increases the evidence required to support superiority claims. Our findings underscore the need for explicit modeling of training variance when validating medical imaging systems. |
| title | Accounting for Underspecification in Statistical Claims of Model Superiority |
| topic | Machine Learning Image and Video Processing |
| url | https://arxiv.org/abs/2511.02453 |