Multifaceted Analysis of Fine-Tuning in a Deep Model for Visual Recognition

Abstract

In recent years, convolutional neural networks (CNNs) have achieved impressive performance for various visual recognition scenarios. CNNs trained on large labeled datasets not only obtain significant performance on most challenging benchmarks but also provide powerful representations, which can be used for a wide range of other tasks. However, the requirement of massive amounts of data to train deep neural networks is a major drawback of these models, as the data available are usually limited or imbalanced. Fine-tuning is an effective way to transfer knowledge learned in a source dataset to a target task. In this article, we introduce and systematically investigate several factors that influence the performance of fine-tuning for visual recognition. These factors include parameters for the retraining procedure (e.g., the initial learning rate of fine-tuning), the distribution of the source and target data (e.g., the number of categories in the source dataset, the distance between the source and target datasets), and so on. We quantitatively and qualitatively analyze these factors, evaluate their influence, and present many empirical observations. The results reveal insights into what fine-tuning changes CNN parameters and provide useful and evidence-backed intuition about how to implement fine-tuning for computer vision tasks.