Now You See Me (CME): Concept-based Model Extraction
This work addresses the need for better interpretability in AI models, particularly for researchers and practitioners in machine learning, though it appears incremental as it builds on existing concept-based analysis methods.
The paper tackles the problem of improving explainability in deep neural networks by introducing CME, a concept-based model extraction framework, which was demonstrated to increase model accuracy by over 14% using only 30% of available concepts in a case study.
Deep Neural Networks (DNNs) have achieved remarkable performance on a range of tasks. A key step to further empowering DNN-based approaches is improving their explainability. In this work we present CME: a concept-based model extraction framework, used for analysing DNN models via concept-based extracted models. Using two case studies (dSprites, and Caltech UCSD Birds), we demonstrate how CME can be used to (i) analyse the concept information learned by a DNN model (ii) analyse how a DNN uses this concept information when predicting output labels (iii) identify key concept information that can further improve DNN predictive performance (for one of the case studies, we showed how model accuracy can be improved by over 14%, using only 30% of the available concepts).