Efficiently Transforming Neural Networks into Decision Trees: A Path to Ground Truth Explanations with RENTT

2025-11-12

AI Summary

Neural networks’ black-box nature impedes interpretability; existing explanation methods suffer from low fidelity, poor accuracy, or limited scalability—especially for large-scale models. Method: We propose RENTT, the first algorithm enabling efficient, precise, and formal translation of neural networks into multivariate decision trees. It supports CNNs, RNNs, diverse activation functions (including non-ReLU), and bias terms, integrating formal transformation theory, multivariate tree construction, and optimized pruning. Contribution/Results: RENTT ensures exact logical equivalence between the original network and its tree representation, enabling global, regional, and local truth-level feature importance analysis. Experiments demonstrate superior computational efficiency and scalability over approximation-based methods (e.g., LIME, SHAP). To our knowledge, RENTT is the only method guaranteeing logically faithful, verifiable, and human-readable explanations. The implementation is publicly available.

Technology Category

• Machine Learning: Ensemble Methods
• Computer Vision: Interpretability, Explainability, and Transparency
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Although neural networks are a powerful tool, their widespread use is hindered by the opacity of their decisions and their black-box nature, which result in a lack of trustworthiness. To alleviate this problem, methods in the field of explainable Artificial Intelligence try to unveil how such automated decisions are made. But explainable AI methods are often plagued by missing faithfulness/correctness, meaning that they sometimes provide explanations that do not align with the neural network's decision and logic. Recently, transformations to decision trees have been proposed to overcome such problems. Unfortunately, they typically lack exactness, scalability, or interpretability as the size of the neural network grows. Thus, we generalize these previous results, especially by considering convolutional neural networks, recurrent neural networks, non-ReLU activation functions, and bias terms. Our findings are accompanied by rigorous proofs and we present a novel algorithm RENTT (Runtime Efficient Network to Tree Transformation) designed to compute an exact equivalent decision tree representation of neural networks in a manner that is both runtime and memory efficient. The resulting decision trees are multivariate and thus, possibly too complex to understand. To alleviate this problem, we also provide a method to calculate the ground truth feature importance for neural networks via the equivalent decision trees - for entire models (global), specific input regions (regional), or single decisions (local). All theoretical results are supported by detailed numerical experiments that emphasize two key aspects: the computational efficiency and scalability of our algorithm, and that only RENTT succeeds in uncovering ground truth explanations compared to conventional approximation methods like LIME and SHAP. All code is available at https://github.com/HelenaM23/RENTT .

Problem

• Overcoming neural network opacity and black-box decision-making limitations
• Addressing faithfulness issues in explainable AI methods for neural networks
• Providing exact and scalable neural network to decision tree transformations

Innovation

• Exact decision tree transformation of neural networks
• Runtime and memory efficient algorithm RENTT
• Ground truth feature importance calculation method