Semi-supervised Image Classification with Grad-CAM Consistency

TL;DR

This paper introduces a semi-supervised image classification method using Grad-CAM consistency loss, improving model accuracy and adaptability across different environments, demonstrated on CIFAR-10 with ResNet.

Contribution

The paper proposes a novel Grad-CAM consistency loss for semi-supervised learning, enhancing model generalization and environment adjustability.

Findings

Achieved up to 1.44% accuracy improvement on CIFAR-10

Demonstrated effectiveness of Grad-CAM consistency over pseudo-labels

Showed adaptability of the method to different model components

Abstract

Consistency training, which exploits both supervised and unsupervised learning with different augmentations on image, is an effective method of utilizing unlabeled data in semi-supervised learning (SSL) manner. Here, we present another version of the method with Grad-CAM consistency loss, so it can be utilized in training model with better generalization and adjustability. We show that our method improved the baseline ResNet model with at most 1.44 % and 0.31 $\pm$ 0.59 %p accuracy improvement on average with CIFAR-10 dataset. We conducted ablation study comparing to using only psuedo-label for consistency training. Also, we argue that our method can adjust in different environments when targeted to different units in the model. The code is available: https://github.com/gimme1dollar/gradcam-consistency-semi-sup.