eXpLogic: Explaining Logic Types and Patterns in DiffLogic Networks

2025-03-13

AI Summary

To address the limited interpretability of logical nodes in DiffLogic networks, this paper proposes eXpLogic: a circuit-inspired method that constructs input-pattern sensitivity graphs to enable node-level logical attribution. It introduces the SwitchDist metric to quantify the discriminative capability of explanation graphs on class-flipping inputs. The framework supports misclassification attribution, discovery of shared activation patterns, and class-specific model compression. Experiments demonstrate that eXpLogic outperforms Vanilla Gradients and Integrated Gradients in explanation fidelity. Under controlled degradation—only 3.8% accuracy loss—it achieves 87% reduction in network size and 8% acceleration in inference speed. This work is the first to integrate logic-gate-level abstraction modeling with gradient-based sensitivity analysis, advancing structured, verifiable, and interpretable deep neural network decision-making research.

Technology Category

• Machine Learning: Probabilistic Circuits and Graphical Models
• Computer Vision: Interpretability, Explainability, and Transparency
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Constraining deep neural networks (DNNs) to learn individual logic types per node, as performed using the DiffLogic network architecture, opens the door to model-specific explanation techniques that quell the complexity inherent to DNNs. Inspired by principles of circuit analysis from computer engineering, this work presents an algorithm (eXpLogic) for producing saliency maps which explain input patterns that activate certain functions. The eXpLogic explanations: (1) show the exact set of inputs responsible for a decision, which helps interpret false negative and false positive predictions, (2) highlight common input patterns that activate certain outputs, and (3) help reduce the network size to improve class-specific inference. To evaluate the eXpLogic saliency map, we introduce a metric that quantifies how much an input changes before switching a model's class prediction (the SwitchDist) and use this metric to compare eXpLogic against the Vanilla Gradients (VG) and Integrated Gradient (IG) methods. Generally, we show that eXpLogic saliency maps are better at predicting which inputs will change the class score. These maps help reduce the network size and inference times by 87% and 8%, respectively, while having a limited impact (-3.8%) on class-specific predictions. The broader value of this work to machine learning is in demonstrating how certain DNN architectures promote explainability, which is relevant to healthcare, defense, and law.

Problem

• Explains input patterns activating specific functions in DNNs.
• Reduces network size and improves class-specific inference.
• Enhances explainability in DNNs for critical applications.

Innovation

• DiffLogic architecture constrains DNNs per node
• eXpLogic algorithm produces interpretable saliency maps
• SwitchDist metric evaluates input change impact