Towards interpretable AI with quantum annealing feature selection

2026-04-28

AI Summary

This work addresses the limited interpretability of convolutional neural networks in image classification by proposing a quantum annealing–based feature map selection method. The approach formulates the identification of critical feature maps as a combinatorial optimization problem and, for the first time, employs quantum annealing to solve it. By integrating quantum constraint encoding with energy gap analysis, the method precisely localizes the basis of model decisions. Experimental results demonstrate that, compared to Grad-CAM and Grad-CAM++, the proposed technique significantly improves explanation quality, enhances class disentanglement, and sharpens decision boundaries. These findings also validate the effectiveness and potential of quantum annealing in advancing explainable artificial intelligence.

Abstract

Deep learning models are used in critical applications, in which mistakes can have serious consequences. Therefore, it is crucial to understand how and why models generate predictions. This understanding provides useful information to check whether the model is learning the right patterns, detect biases in the data, improve model design, and build systems that can be trusted. This work proposes a new method for interpreting Convolutional Neural Networks in image classification tasks. The approach works by selecting the most important feature maps that contribute to each prediction. To solve this combinatorial problem, we encode it into a quantum constrained optimization problem and propose to solve it using quantum annealing. We evaluate our method against the state-of-the-art explainable AI techniques, specifically GradCAM and GradCAM++, and observe an improved class disentanglement, i.e. the model's decision boundaries become more distinct and its reasoning more transparent. This demonstrates that our approach enhances the quality of explanations, making it easier to understand which features the model relies on for specific predictions. In addition, we study the computational behavior of the quantum annealing algorithm. Specifically, we analyze the minimum energy gap of the system during computation and the probability that the algorithm finds the correct solution. These analyses provide theoretical insight into why the method works effectively in practice.

Problem

• interpretable AI
• feature selection
• Convolutional Neural Networks
• explainability
• image classification

Innovation

• quantum annealing
• feature selection
• interpretable AI
• convolutional neural networks
• combinatorial optimization