Using LRP scores, the method prunes negative-relevance filters and quantizes remaining weights by per-layer median, yielding 90.5% accuracy at 1.2 MB on a synthetic 4-class task.
Explaining deep neural networks and beyond: A review of methods and applications
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Compressing Deep Neural Networks Using Explainable AI
Using LRP scores, the method prunes negative-relevance filters and quantizes remaining weights by per-layer median, yielding 90.5% accuracy at 1.2 MB on a synthetic 4-class task.