Neuron Incidence Redistribution reduces demographic performance gaps in medical image classifiers by redistributing latent disease evidence more evenly across the penultimate layer without using demographic labels.
FairALM: Augmented Lagrangian method for training fair models with little regret,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Neuron Incidence Redistribution for Fairness in Medical Image Classification
Neuron Incidence Redistribution reduces demographic performance gaps in medical image classifiers by redistributing latent disease evidence more evenly across the penultimate layer without using demographic labels.