ExplainabilityTransparencyAccountabilitycs.LG
A Novel Approach to Explainable AI with Quantized Active Ingredients in Decision Making
A. M. A. S. D. Alagiyawanna, Asoka Karunananda, Thushari Silva, A. Mahasinghe
January 13, 2026
2025 9th SLAAI International Conference on Artificial Intelligence (SLAAI-ICAI)
Key Takeaways
-
Explainability is critical for high-stakes AI decisions: The paper emphasizes that AI systems used in healthcare and finance must be transparent about how they reach conclusions, not just accurate—a key principle for responsible AI deployment in sensitive domains.
-
Quantum-enhanced AI can improve both accuracy and interpretability: The study found that Quantum Boltzmann Machines achieved 83.5% classification accuracy compared to 54% for classical models, while also providing clearer identification of which features (or "active ingredients") drove their decisions, suggesting quantum approaches may offer dual benefits.
-
"Active ingredients" matter for accountability: By using techniques like SHAP and saliency maps to identify the most important factors in AI decisions, the framework makes it possible to understand and potentially challenge why an AI system made a particular prediction—essential for responsible AI in regulated industries.
-
Clearer feature attribution reduces ambiguity: The paper measured interpretability using entropy scores, finding QBMs produced more concentrated distributions (1.27 vs. 1.39), meaning the models pointed to a clearer set of decisive factors rather than spreading importance across many features—making explanations more actionable and trustworthy.
Abstract
Artificial Intelligence (AI) systems have shown good success at classifying. However, the lack of explainability is a true and significant challenge, especially in high-stakes domains, such as health and finance, where understanding is paramount. We propose a new solution to this challenge: an explainable AI framework based on our comparative study with Quantum Boltzmann Machines (QBMs) and Classical Boltzmann Machines (CBMs). We leverage principles of quantum computing within classical machine learning to provide substantive transparency around decision-making. The design involves training both models on a binarised and dimensionally reduced MNIST dataset, where Principal Component Analysis (PCA) is applied for preprocessing. For interpretability, we employ gradient-based saliency maps in QBMs and SHAP (SHapley Additive exPlanations) in CBMs to evaluate feature attributions.QBMs deploy hybrid quantum-classical circuits with strongly entangling layers, allowing for richer latent representations, whereas CBMs serve as a classical baseline that utilises contrastive divergence. Along the way, we found that QBMs outperformed CBMs on classification accuracy (83.5% vs. 54%) and had more concentrated distributions in feature attributions as quantified by entropy (1.27 vs. 1.39). In other words, QBMs not only produced better predictive performance than CBMs, but they also provided clearer identification of "active ingredient" or the most important features behind model predictions. To conclude, our results illustrate that quantum-classical hybrid models can display improvements in both accuracy and interpretability, which leads us toward more trustworthy and explainable AI systems.