Abstract:The rapid evolution of deep learning has led to significant achievements in computer vision, primarily driven by complex convolutional neural networks (CNNs). However, the increasing depth and parameter count of these networks often result in overfitting and elevated computational demands. Knowledge distillation (KD) has emerged as a promising technique to address these issues by transferring knowledge from a large, well-trained teacher model to a more compact student model. This paper introduces a novel knowledge distillation method that simplifies the distillation process and narrows the performance gap between teacher and student models without relying on intricate knowledge representations. Our approach leverages a unique teacher network architecture designed to enhance the efficiency and effectiveness of knowledge transfer. Additionally, we introduce a streamlined teacher network architecture that transfers knowledge effectively through a simplified distillation process, enabling the student model to achieve high accuracy with reduced computational demands. Comprehensive experiments conducted on the CIFAR-10 dataset demonstrate that our proposed model achieves superior performance compared to traditional KD methods and established architectures such as ResNet and VGG networks. The proposed method not only maintains high accuracy but also significantly reduces training and validation losses. Key findings highlight the optimal hyperparameter settings (temperature T = 15.0 and smoothing factor α = 0.7), which yield the highest validation accuracy and lowest loss values. This research contributes to the theoretical and practical advancements in knowledge distillation, providing a robust framework for future applications and research in neural network compression and optimization. The simplicity and efficiency of our approach pave the way for more accessible and scalable solutions in deep learning model deployment.

Few Sample Knowledge Distillation for Efficient Network Compression

DCCD: Reducing Neural Network Redundancy Via Distillation

An Efficient Method for Model Pruning Using Knowledge Distillation with Few Samples.

Pruning by Training: A Novel Deep Neural Network Compression Framework for Image Processing.

A Model Compression Method Using Significant Data and Knowledge Distillation

Using Distillation to Improve Network Performance after Pruning and Quantization

Progressive Network Grafting for Few-Shot Knowledge Distillation

Few Shot Network Compression via Cross Distillation

Triplet Knowledge Distillation Networks for Model Compression.

Knowledge from the Original Network: Restore a Better Pruned Network with Knowledge Distillation

Analysis of Model Compression Using Knowledge Distillation

On Compressing U-net Using Knowledge Distillation

Towards Efficient Network Compression Via Few-Shot Slimming.

Pruning-and-distillation: One-stage Joint Compression Framework for CNNs Via Clustering

A Novel Deep Learning Model Compression Algorithm

Hierarchical Knowledge Squeezed Adversarial Network Compression

Parallel Blockwise Knowledge Distillation for Deep Neural Network Compression

Simplified Knowledge Distillation for Deep Neural Networks Bridging the Performance Gap with a Novel Teacher–Student Architecture

A Novel Architecture Slimming Method for Network Pruning and Knowledge Distillation

ResKD: Residual-Guided Knowledge Distillation

Model Compression Algorithm Via Reinforcement Learning and Knowledge Distillation