MICCAI 2026 Tutorial to Explain Zeta Scaling Law Behind Medical AI Competition Rankings

PTenigma · x · 2026-09-21

A tutorial at MICCAI 2026 (Oct 1, will be recorded) will explain the math behind medical AI competition leaderboards: learning curves and confidence cones showing how each algorithm's mean AUC and bootstrap spread evolve with data, why higher-capacity models "cross over" simpler ones at a critical sample size, and the zeta scaling law, which models disease signals across embedding eigenvector directions via Davis-Kahan/Weyl theorems.

Original post →

More from Research

Research channel →