You can check below one of our recently published articles in IEEE Transactions on Artificial Intelligence , entitled “Exploring Energy Landscapes for Minimal Counterfactual Explanations: Applications in Cybersecurity and Beyond” and authored by
Spyros Evangelatos, Eleni Veroni, Vasilis Efthymiou, Christos D. Nikolopoulos, Georgios Th. Papadopoulos and Panagiotis Sarigiannidis
đź’» Paper link: https://ieeexplore.ieee.org/document/11199968
In this work, we present a novel framework that integrates perturbation theory and statistical mechanics to generate minimal counterfactual explanations in explainable AI. We employ a local Taylor expansion of a Machine Learning model’s predictive function and reformulate the counterfactual search as an energy minimization problem over a complex landscape. In sequence, we model the probability of candidate perturbations leveraging the Boltzmann distribution and use simulated annealing for iterative refinement. Our approach systematically identifies the smallest modifications required to change a model’s prediction while maintaining plausibility.
This paper comprises a scientific outcome of the European RnD security project GANNDALF
