Functional Rule Extraction Method for Artificial Neural Networks
This addresses the need for interpretability in neural networks for researchers and practitioners, but appears incremental as it builds on existing rule extraction concepts.
The paper tackles the problem of extracting interpretable rules from artificial neural networks by proposing a method based on comprehensive functions for directed and undirected rule extraction, but it does not provide concrete results or numbers.
The idea I propose in this paper is a method that is based on comprehensive functions for directed and undirected rule extraction from artificial neural network operations. Firstly, I defined comprehensive functions, then constructed a comprehensive multilayer network (denoted as N). Each activation function of N is parametrized to a comprehensive function.