Quantifying the Knowledge in Deep Neural Networks: An Overview
Ioanna Valsamara, Ioannis Mademlis, Ioannis Pitas · ACM Computing Surveys · 2026
Deep Neural Networks (DNNs) have proven to be extremely effective at learning a wide range of tasks. Due to their complexity and inexplicable internal state, DNNs are difficult to analyze. Several attempts to interpret their operation have been made, but analyzing them from the perspective of the knowledge encoded in their layers is a promising research direction. The purpose of this survey is two-fold: (a) to review the concept of DNN knowledge quantification and highlight it as an important near-future challenge, and (b) to provide a brief account of the scant existing methods attempting to actually quantify DNN knowledge.