Decoding Interpretable Logic Rules from Neural Networks

2025-01-14

AI Summary

Deep neural networks (DNNs) face deployment barriers in high-stakes domains due to their opaque decision-making, especially for complex architectures like ResNet and Transformers. To address this, we propose NeuroLogic—a novel method that automatically extracts globally coherent, human-interpretable symbolic logic rules from large-scale vision models. NeuroLogic integrates neural activation pattern analysis with visual concept grounding to construct a differentiable logical reasoning framework based on latent predicates, supporting hierarchical interpretation across diverse architectures including CNNs and Transformers. Experiments demonstrate that NeuroLogic achieves over 85% rule coverage on multiple benchmarks, with high fidelity and formal verifiability—substantially outperforming state-of-the-art explanation techniques. By bridging neural computation and symbolic logic, NeuroLogic establishes a new paradigm for trustworthy DNN deployment.

Technology Category

• Computer Vision: Visual Reasoning & Symbolic Representations
• Machine Learning: Neuro-Symbolic Learning
• Knowledge Representation and Reasoning: Logic Programming

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

As deep neural networks continue to excel across various domains, their black-box nature has raised concerns about transparency and trust. In particular, interpretability has become increasingly essential for applications that demand high safety and knowledge rigor, such as drug discovery, autonomous driving, and genomics. However, progress in understanding even the simplest deep neural networks - such as fully connected networks - has been limited, despite their role as foundational elements in state-of-the-art models like ResNet and Transformer. In this paper, we address this challenge by introducing NeuroLogic, a novel approach for decoding interpretable logic rules from neural networks. NeuroLogic leverages neural activation patterns to capture the model's critical decision-making processes, translating them into logical rules represented by hidden predicates. Thanks to its flexible design in the grounding phase, NeuroLogic can be adapted to a wide range of neural networks. For simple fully connected neural networks, hidden predicates can be grounded in certain split patterns of original input features to derive decision-tree-like rules. For large, complex vision neural networks, NeuroLogic grounds hidden predicates into high-level visual concepts that are understandable to humans. Our empirical study demonstrates that NeuroLogic can extract global and interpretable rules from state-of-the-art models such as ResNet, a task at which existing work struggles. We believe NeuroLogic can help pave the way for understanding the black-box nature of neural networks.

Problem

• Interpretable AI
• Deep Neural Networks
• Explainability in ML

Innovation

• NeuroLogic
• Interpretable Rules
• Neural Network Transparency