Integrating 15% logistic chaos perturbations into prototypical networks raises 4-way 5-shot brain tumor accuracy to 84.52% by producing more stable, noise-invariant embeddings.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.IV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Chaos-Enhanced Prototypical Networks for Few-Shot Medical Image Classification
Integrating 15% logistic chaos perturbations into prototypical networks raises 4-way 5-shot brain tumor accuracy to 84.52% by producing more stable, noise-invariant embeddings.