Far from training data, ReLU classifiers grow more confident with every step — a 2019 theorem explains why softmax can't say 'none of these'

bravo_abad · x · 2026-11-23

Jorge Bravo Abad explains a 2019 theorem by Hein, Andriushchenko and Bitterwolf: ReLU classifiers, given inputs not confined to a fixed range, see confidence tend to 100% far from training data.

Why

Consequences

Softmax ranks the known answers; it can't tell you the question is new.

Original post →

More from Research

Research channel →