Symbiotic Hybrid Neural Network Watchdog For Outlier Detection
Machine Learning
2021-09-14 v2 Image and Video Processing
Abstract
Neural networks are largely black boxes. A neural network trained to classify fruit may classify a picture of a giraffe as a banana. A neural network watchdog's job is to identify such inputs, allowing a classifier to disregard such data. We investigate whether the watchdog should be separate from the neural network or symbiotically attached. We present empirical evidence that the symbiotic watchdog performs better than when the neural networks are disjoint.
Cite
@article{arxiv.2103.00582,
title = {Symbiotic Hybrid Neural Network Watchdog For Outlier Detection},
author = {Justin Bui and Robert J. Marks},
journal= {arXiv preprint arXiv:2103.00582},
year = {2021}
}
Comments
10 Pages, 25 References, 10 Figures, 5 Tables