RT Journal Article
T1 Greybox XAI: 												A Neural-Symbolic learning framework to produce interpretable predictions for image classification
A1 Bennetot, Adrien
A1 Franchi, Gianni
A1 Ser, Javier Del
A1 Chatila, Raja
A1 Díaz-Rodríguez, Natalia
AB Although Deep Neural Networks (DNNs) have great generalization and prediction capabilities, their functioning does not allow a detailed explanation of their behavior. Opaque deep learning models are increasingly used to make important predictions in critical environments, and the danger is that they make and use predictions that cannot be justified or legitimized. Several eXplainable Artificial Intelligence (XAI) methods that separate explanations from machine learning models have emerged, but have shortcomings in faithfulness to the model actual functioning and robustness. As a result, there is a widespread agreement on the importance of endowing Deep Learning models with explanatory capabilities so that they can themselves provide an answer to why a particular prediction was made. First, we address the problem of the lack of universal criteria for XAI by formalizing what an explanation is. We also introduced a set of axioms and definitions to clarify XAI from a mathematical perspective. Finally, we present the Greybox XAI, a framework that composes a DNN and a transparent model thanks to the use of a symbolic Knowledge Base (KB). We extract a KB from the dataset and use it to train a transparent model (i.e., a logistic regression). An encoder–decoder architecture is trained on RGB images to produce an output similar to the KB used by the transparent model. Once the two models are trained independently, they are used compositionally to form an explainable predictive model. We show how this new architecture is accurate and explainable in several datasets.
SN 0950-7051
YR 2022
FD 2022-12-22
LK https://hdl.handle.net/11556/3617
UL https://hdl.handle.net/11556/3617
LA eng
NO Bennetot , A , Franchi , G , Ser , J D , Chatila , R & Díaz-Rodríguez , N 2022 , ' Greybox XAI : A Neural-Symbolic learning framework to produce interpretable predictions for image classification ' , Knowledge-Based Systems , vol. 258 , 109947 . https://doi.org/10.1016/j.knosys.2022.109947
NO Publisher Copyright: © 2022 Elsevier B.V.
NO This research was funded by the French ANRT (Association Nationale Recherche Technologie — ANRT) industrial CIFRE PhD contract with SEGULA Technologies. Díaz-Rodríguez is supported by Juan de la Cierva Incorporación grant IJC2019-039152-I funded by MCIN/AEI /10.13039/501100011033 by “ ESF Investing in your future” and Google Research Scholar Program . J. Del Ser acknowledges support from the Department of Education of the Basque Government (Consolidated Research Group MATHMODE, IT1456-22 ).
DS TECNALIA Publications
RD 31 jul 2024