A Methodology for Transparent Logic-Based Classification Using a Multi-Task Convolutional Tsetlin Machine

2025-10-02

AI Summary

This work addresses the challenge of balancing interpretability and performance in large-scale, multi-channel RGB image classification. We propose a novel multi-task Convolutional Tsetlin Machine (TM) architecture grounded in finite-state automata and propositional logic, jointly learning local predictive explanations (clause-level) and global class representations (class-level logical patterns), with convolutional clauses visualized to enhance transparency. Our approach is the first to achieve both high interpretability and competitive accuracy on complex real-world data (CelebA). It achieves 98.5% accuracy on MNIST and an F1-score of 86.56% on CelebA—comparable to ResNet50—while preserving full logical transparency. This work establishes a scalable, interpretable, and practically viable paradigm for logic-driven transparent AI in image classification.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Machine Learning: Transparent, Interpretable, Explainable ML
• Knowledge Representation and Reasoning: Logic Programming

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Data transparency and provenance

Abstract

The Tsetlin Machine (TM) is a novel machine learning paradigm that employs finite-state automata for learning and utilizes propositional logic to represent patterns. Due to its simplistic approach, TMs are inherently more interpretable than learning algorithms based on Neural Networks. The Convolutional TM has shown comparable performance on various datasets such as MNIST, K-MNIST, F-MNIST and CIFAR-2. In this paper, we explore the applicability of the TM architecture for large-scale multi-channel (RGB) image classification. We propose a methodology to generate both local interpretations and global class representations. The local interpretations can be used to explain the model predictions while the global class representations aggregate important patterns for each class. These interpretations summarize the knowledge captured by the convolutional clauses, which can be visualized as images. We evaluate our methods on MNIST and CelebA datasets, using models that achieve 98.5% accuracy on MNIST and 86.56% F1-score on CelebA (compared to 88.07% for ResNet50) respectively. We show that the TM performs competitively to this deep learning model while maintaining its interpretability, even in large-scale complex training environments. This contributes to a better understanding of TM clauses and provides insights into how these models can be applied to more complex and diverse datasets.

Problem

• Developing transparent logic-based classification using multi-task convolutional Tsetlin Machine
• Generating local interpretations and global representations for model predictions
• Maintaining interpretability while achieving competitive performance on complex datasets

Innovation

• Multi-task convolutional Tsetlin Machine for classification
• Propositional logic representation for interpretable patterns
• Local and global visual explanations from clauses