The Tsetlin Machine Goes Deep: Logical Learning and Reasoning With Graphs
This work addresses the need for interpretable and efficient graph learning methods in fields like computer vision, natural language processing, and bioinformatics, offering a novel approach that is not incremental but introduces a new method for a known bottleneck.
The paper tackles the problem of learning from graph-structured data by introducing the Graph Tsetlin Machine (GraphTM), which uses message passing to build nested deep clauses for recognizing sub-graph patterns, resulting in improved accuracy across tasks such as image classification (3.86%-points higher on CIFAR-10), action coreference tracking (up to 20.6%-points better), and recommendation systems (89.86% vs. 70.87% accuracy under noise).
Pattern recognition with concise and flat AND-rules makes the Tsetlin Machine (TM) both interpretable and efficient, while the power of Tsetlin automata enables accuracy comparable to deep learning on an increasing number of datasets. We introduce the Graph Tsetlin Machine (GraphTM) for learning interpretable deep clauses from graph-structured input. Moving beyond flat, fixed-length input, the GraphTM gets more versatile, supporting sequences, grids, relations, and multimodality. Through message passing, the GraphTM builds nested deep clauses to recognize sub-graph patterns with exponentially fewer clauses, increasing both interpretability and data utilization. For image classification, GraphTM preserves interpretability and achieves 3.86%-points higher accuracy on CIFAR-10 than a convolutional TM. For tracking action coreference, faced with increasingly challenging tasks, GraphTM outperforms other reinforcement learning methods by up to 20.6%-points. In recommendation systems, it tolerates increasing noise to a greater extent than a Graph Convolutional Neural Network (GCN), e.g., for noise ratio 0.1, GraphTM obtains accuracy 89.86% compared to GCN's 70.87%. Finally, for viral genome sequence data, GraphTM is competitive with BiLSTM-CNN and GCN accuracy-wise, training 2.5x faster than GCN. The GraphTM's application to these varied fields demonstrates how graph representation learning and deep clauses bring new possibilities for TM learning.