Uncertainty-Aware Dual-Student Knowledge Distillation for Efficient Image Classification

TL;DR

This paper introduces an uncertainty-aware dual-student knowledge distillation method that improves image classification by leveraging teacher confidence and collaborative learning between heterogeneous student models.

Contribution

It presents a novel framework combining uncertainty estimation with peer learning of two different student architectures for enhanced model compression.

Findings

ResNet-18 achieves 83.84% top-1 accuracy

MobileNetV2 achieves 81.46% top-1 accuracy

Outperforms traditional distillation methods on ImageNet-100

Abstract

Knowledge distillation has emerged as a powerful technique for model compression, enabling the transfer of knowledge from large teacher networks to compact student models. However, traditional knowledge distillation methods treat all teacher predictions equally, regardless of the teacher's confidence in those predictions. This paper proposes an uncertainty-aware dual-student knowledge distillation framework that leverages teacher prediction uncertainty to selectively guide student learning. We introduce a peer-learning mechanism where two heterogeneous student architectures, specifically ResNet-18 and MobileNetV2, learn collaboratively from both the teacher network and each other. Experimental results on ImageNet-100 demonstrate that our approach achieves superior performance compared to baseline knowledge distillation methods, with ResNet-18 achieving 83.84\% top-1 accuracy and MobileNetV2 achieving 81.46\% top-1 accuracy, representing improvements of 2.04\% and 0.92\% respectively over traditional single-student distillation approaches.