Pruning by explaining: A novel criterion for deep neural network pruning

• Published in: Pattern recognition Vol. 115; p. 107899
• Main Authors: Yeom, Seul-Ki, Seegerer, Philipp, Lapuschkin, Sebastian, Binder, Alexander, Wiedemann, Simon, Müller, Klaus-Robert, Samek, Wojciech
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.07.2021
• Subjects: Convolutional neural network (CNN); Explainable AI (XAI); Interpretation of models; Layer-wise relevance propagation (LRP); Pruning; Pruning; Interpretation of models; Convolutional neural network (CNN); Layer-wise relevance propagation (LRP); Explainable AI (XAI)

•A novel criterion to efficiently prune convolutional neural networks inspired by explaining nonlinear classification decisions in terms of input variables is introduced.•The method is inspired by neural network interpretability: Layer-wise Relevance Propagation.•This is the first report to link the...