A Self-Explainable Deep Architecture for Security Applications

2026-08-05

AI Summary

This work addresses the limited interpretability of deep learning models in security applications, where existing post-hoc explanation methods often suffer from low fidelity, instability, and high computational overhead. To overcome these challenges, the paper introduces XSec—the first intrinsically interpretable deep architecture tailored for the security domain. XSec integrates mask-based feature selection and class-prototype learning during training, coupled with a similarity-measurement layer at inference time, thereby enabling end-to-end interpretability without reliance on post-hoc analysis. Evaluated across five security tasks, XSec achieves an average accuracy of 97.33% while delivering explanations that are highly faithful, stable, and low-latency. This study marks the first successful integration of self-explainable AI into security applications, significantly enhancing model transparency and reliability without compromising performance.

Abstract

Deep learning models have become integral to security applications due to their ability to model complex relationships in data and detect sophisticated threats. However, their complexity makes it difficult to understand how predictions are generated, posing significant challenges for interpretability, particularly in security applications where transparency is critical. Existing explanation methods, such as visual explanation techniques and post-hoc approaches, suffer from several limitations: reduced faithfulness due to local approximation errors, instability caused by reliance on randomness, and computational inefficiency that hinders real-time usage. To address these issues, we introduce XSec, a self-explainable deep architecture developed for security applications. During training, XSec uses a novel mask-based approach to extract informative sub-features from the data and learns prototypes, representative patterns that characterize each class. XSec then leverages the prototypes in a dedicated similarity layer at test time to compute similarity scores and generates interpretable explanations without the need for post-hoc analysis. We evaluate XSec across five diverse security scenarios, demonstrating its ability to achieve an average classification accuracy of 97.33% with minimal performance compromise. XSec produces deterministic explanations for a fixed trained model and input and substantially reduces explanation latency compared with approximation-based and perturbation-based post-hoc methods. Through this effort, we extend the applicability of self-explainable AI to security applications, bridging the gap between deep learning performance and the need for explainability in critical scenarios.

Problem

• explainability
• security applications
• deep learning
• interpretability
• post-hoc methods

Innovation

• self-explainable AI
• prototype learning
• mask-based feature extraction
• similarity layer
• security applications