The use of transfer learning with deep neural networks has increasingly become widespread for deploying well-tested computer vision systems to newer domains, especially those with limited datasets. We describe a transfer learning use case for a domain with a data-starved regime, having fewer than 100 labeled target samples. We evaluate the effectiveness of convolutional feature extraction and fine-tuning of overparameterized models with respect to the size of target training data, as well as their generalization performance on data with covariate shift, or out-of-distribution (OOD) data. Our experiments demonstrate that both overparameterization and feature reuse contribute to the successful application of transfer learning in training image classifiers in data-starved regimes. We provide visual explanations to support our findings and conclude that transfer learning enhances the performance of CNN architectures in data-starved regimes.
@article{arxiv.2003.04117,
title = {The Utility of Feature Reuse: Transfer Learning in Data-Starved Regimes},
author = {Rashik Shadman and M. G. Sarwar Murshed and Edward Verenich and Alvaro Velasquez and Faraz Hussain},
journal= {arXiv preprint arXiv:2003.04117},
year = {2023}
}