Comparison of Machine Learning Approaches for Classifying Spinodal Events

TL;DR

This work evaluates state-of-the-art models (MobileViT, NAT, EfficientNet, CNN), alongside several ensemble models (majority voting, AdaBoost), and explores the dataset in a transformed color space.

Abstract

In this work, we compare the performance of deep learning models for classifying the spinodal dataset. We evaluate state-of-the-art models (MobileViT, NAT, EfficientNet, CNN), alongside several ensemble models (majority voting, AdaBoost). Additionally, we explore the dataset in a transformed color space. Our findings show that NAT and MobileViT outperform other models, achieving the highest metrics-accuracy, AUC, and F1 score on both training and testing data (NAT: 94.65, 0.98, 0.94; MobileViT: 94.20, 0.98, 0.94), surpassing the earlier CNN model (88.44, 0.95, 0.88). We also discuss failure cases for the top performing models.

Figures (8)

• Figure 1: Spinodal and Non-Spinodal events in the dataset.
• Figure 2: Architecture diagram of the CNN network.
• Figure 3: Majority voting model
• Figure 4: Loss as a function of number of epochs on the training dataset.
• Figure 5: Accuracy of models as a function of number of epochs on the training dataset.