A small MNIST classifier is far more sensitive to shifts of internal decision boundaries than to scaling or clipping of activations, supporting the author's claim that such networks use distance-like representations.
Encyclopedia of distances
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Neural Networks Use Distance Metrics
A small MNIST classifier is far more sensitive to shifts of internal decision boundaries than to scaling or clipping of activations, supporting the author's claim that such networks use distance-like representations.