Native Explainability for Bayesian Confidence Propagation Neural Networks: A Framework for Trusted Brain-Like AI
For researchers and practitioners deploying brain-like AI on edge devices, this work fills a critical gap by enabling native explainability in BCPNNs, which is essential for regulatory compliance and trust in high-risk AI systems.
The paper proposes the first systematic explainability framework for Bayesian Confidence Propagation Neural Networks (BCPNN), introducing a taxonomy, 16 architecture-level explanation primitives, and 5 design-time configuration primitives to make BCPNN decisions transparent and aligned with the EU AI Act. The framework leverages BCPNN's inherent interpretability to provide closed-form algorithms for attribution, prototype, concept, counterfactual, and mechanistic explanations.
The EU Artificial Intelligence Act (Regulation 2024/1689), fully applicable to high-risk systems from August 2026, creates urgent demand for AI architectures that are simultaneously trustworthy, transparent, and feasible to deploy on resource-constrained edge devices. Brain-like neural networks built on the Bayesian Confidence Propagation Neural Network (BCPNN) formalism have re-emerged as a credible alternative to backpropagation-driven deep learning. They deliver state-of-the-art unsupervised representation learning, neuromorphic-friendly sparsity, and existing FPGA implementations that target edge deployment. Despite this momentum, no systematic framework exists for explaining BCPNN decisions -- a gap the present paper fills. We argue that BCPNN is, in the sense of Rudin's interpretable-by-design agenda, an inherently transparent model whose architectural primitives map directly onto established explainable-AI (XAI) families. We make four contributions. First, we propose the first XAI taxonomy for BCPNN. It maps weights, biases, hypercolumn posteriors, structural-plasticity usage scores, attractor dynamics, and input-reconstruction populations onto attribution, prototype, concept, counterfactual, and mechanistic explanation modalities. Second, we introduce sixteen architecture-level explanation primitives (P1--P16), several without analogue in standard ANNs. We provide closed-form algorithms for computing each from quantities the model already maintains. Third, we introduce five design-time Configuration-as-Explanation primitives (Config-P1 to Config-P5) that treat BCPNN hyperparameter choices as an auditable pre-deployment explanation artifact. Fourth, we sketch a roadmap for integration into industrial IoT deployments and discuss EU AI Act alignment, edge feasibility, and Industry 5.0 implications.