Inspect Transfer Learning Architecture with Dilated Convolution

TL;DR

This paper explores architectural modifications to pre-trained VGG networks, including freezing early layers and applying dilated convolutions, to improve image classification accuracy on CIFAR datasets.

Contribution

The paper introduces modified VGG architectures with dilated convolutions and layer freezing, enhancing transfer learning performance for image classification.

Findings

Modified architectures outperform baseline VGG models on CIFAR datasets.

Freezing early convolutional layers reduces overfitting.

Dilated convolutions improve feature extraction at lower resolutions.

Abstract

There are many award-winning pre-trained Convolutional Neural Network (CNN), which have a common phenomenon of increasing depth in convolutional layers. However, I inspect on VGG network, which is one of the famous model submitted to ILSVRC-2014, to show that slight modification in the basic architecture can enhance the accuracy result of the image classification task. In this paper, We present two improve architectures of pre-trained VGG-16 and VGG-19 networks that apply transfer learning when trained on a different dataset. I report a series of experimental result on various modification of the primary VGG networks and achieved significant out-performance on image classification task by: (1) freezing the first two blocks of the convolutional layers to prevent over-fitting and (2) applying different combination of dilation rate in the last three blocks of convolutional layer to reduce image resolution for feature extraction. Both the proposed architecture achieves a competitive result on CIFAR-10 and CIFAR-100 dataset.