An Experimental Study of the Impact of Pre-training on the Pruning of a Convolutional Neural Network

Abstract

In recent years, deep neural networks have known a wide success in various application domains. However, they require important computational and memory resources, which severely hinders their deployment, notably on mobile devices or for real-time applications. Neural networks usually involve a large number of parameters, which correspond to the weights of the network. Such parameters, obtained with the help of a training process, are determinant for the performance of the network. However, they are also highly redundant. The pruning methods notably attempt to reduce the size of the parameter set, by identifying and removing the irrelevant weights. In this paper, we examine the impact of the training strategy on the pruning efficiency. Two training modalities are considered and compared: (1) fine-tuned and (2) from scratch. The experimental results obtained on four datasets (CIFAR10, CIFAR100, SVHN and Caltech101) and for two different CNNs (VGG16 and MobileNet) demonstrate that a network that has been pre-trained on a large corpus (e.g. ImageNet) and then fine-tuned on a particular dataset can be pruned much more efficiently (up to 80% of parameter reduction) than the same network trained from scratch.

arXiv comment7 pages, published at APPIS 2020

References in corpus (11)

• Distilling the Knowledge in a Neural Network
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Do Deep Nets Really Need to be Deep?
• Pruning Filters for Efficient ConvNets
• Xception: Deep Learning with Depthwise Separable Convolutions
• Channel Pruning for Accelerating Very Deep Neural Networks
• XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks
• One ticket to win them all: generalizing lottery ticket initializations across datasets and optimizers
• Structured Pruning of Deep Convolutional Neural Networks
• Convolutional Neural Networks at Constrained Time Cost
• Recent Advances in Efficient Computation of Deep Convolutional Neural Networks

Cited by in corpus (2)

• Induced Feature Selection by Structured Pruning
• One-Cycle Pruning: Pruning ConvNets Under a Tight Training Budget