Entropy-based Logic Explanations of Neural Networks

Abstract

Explainable artificial intelligence has rapidly emerged since lawmakers have started requiring interpretable models for safety-critical domains. Concept-based neural networks have arisen as explainable-by-design methods as they leverage human-understandable symbols (i.e. concepts) to predict class memberships. However, most of these approaches focus on the identification of the most relevant concepts but do not provide concise, formal explanations of how such concepts are leveraged by the classifier to make predictions. In this paper, we propose a novel end-to-end differentiable approach enabling the extraction of logic explanations from neural networks using the formalism of First-Order Logic. The method relies on an entropy-based criterion which automatically identifies the most relevant concepts. We consider four different case studies to demonstrate that: (i) this entropy-based criterion enables the distillation of concise logic explanations in safety-critical domains from clinical data to computer vision; (ii) the proposed approach outperforms state-of-the-art white-box models in terms of classification accuracy and matches black box performances.

References in corpus (6)

• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Towards A Rigorous Science of Interpretable Machine Learning
• Opportunities and Challenges in Explainable Artificial Intelligence (XAI): A Survey
• Learning Symbolic Physics with Graph Networks
• Now You See Me (CME): Concept-based Model Extraction

Cited by in corpus (9)

• Concept-based Explainable Artificial Intelligence: A Survey
• DeepCover: Advancing RNN Test Coverage and Online Error Prediction using State Machine Extraction
• A Comprehensive Survey on Self-Interpretable Neural Networks
• Extending Logic Explained Networks to Text Classification
• E Pluribus Unum Interpretable Convolutional Neural Networks
• PyTorch, Explain! A Python library for Logic Explained Networks
• Understanding Post-hoc Explainers: The Case of Anchors
• Cause and Effect: Hierarchical Concept-based Explanation of Neural Networks
• It Ain't That Bad: Understanding the Mysterious Performance Drop in OOD Generalization for Generative Transformer Models