An Optimized Toolbox for Advanced Image Processing with Tsetlin Machine Composites

TL;DR

This paper introduces a new toolbox of specialized Tsetlin Machine components using advanced image processing techniques, achieving state-of-the-art accuracy on CIFAR-10 and advancing TM-based image classification.

Contribution

The paper presents a novel TM Composites architecture with specialized image processing techniques and optimized hyperparameters, significantly improving CIFAR-10 accuracy.

Findings

Achieved 82.8% accuracy on CIFAR-10 with TM Composites.

Developed a set of TM Specialists using various image processing methods.

Provided a hyperparameter search that identified optimal settings for TM Specialists.

Abstract

The Tsetlin Machine (TM) has achieved competitive results on several image classification benchmarks, including MNIST, K-MNIST, F-MNIST, and CIFAR-2. However, color image classification is arguably still in its infancy for TMs, with CIFAR-10 being a focal point for tracking progress. Over the past few years, TM's CIFAR-10 accuracy has increased from around 61% in 2020 to 75.1% in 2023 with the introduction of Drop Clause. In this paper, we leverage the recently proposed TM Composites architecture and introduce a range of TM Specialists that use various image processing techniques. These include Canny edge detection, Histogram of Oriented Gradients, adaptive mean thresholding, adaptive Gaussian thresholding, Otsu's thresholding, color thermometers, and adaptive color thermometers. In addition, we conduct a rigorous hyperparameter search, where we uncover optimal hyperparameters for several of the TM Specialists. The result is a toolbox that provides new state-of-the-art results on CIFAR-10 for TMs with an accuracy of 82.8%. In conclusion, our toolbox of TM Specialists forms a foundation for new TM applications and a landmark for further research on TM Composites in image analysis.