Bridging the Gap: Simultaneous Fine Tuning for Data Re-Balancing
Abstract
There are many real-world classification problems wherein the issue of data imbalance (the case when a data set contains substantially more samples for one/many classes than the rest) is unavoidable. While under-sampling the problematic classes is a common solution, this is not a compelling option when the large data class is itself diverse and/or the limited data class is especially small. We suggest a strategy based on recent work concerning limited data problems which utilizes a supplemental set of images with similar properties to the limited data class to aid in the training of a neural network. We show results for our model against other typical methods on a real-world synthetic aperture sonar data set. Code can be found at github.com/JohnMcKay/dataImbalance.
Cite
@article{arxiv.1801.02548,
title = {Bridging the Gap: Simultaneous Fine Tuning for Data Re-Balancing},
author = {John McKay and Isaac Gerg and Vishal Monga},
journal= {arXiv preprint arXiv:1801.02548},
year = {2018}
}
Comments
Submitted to IGARSS 2018, 4 Pages, 8 Figures