Autoencoder Watchdog Outlier Detection for Classifiers

Justin Bui, Robert Marks II

2021

Abstract

Neural networks have often been described as black boxes. A generic neural network trained to differentiate between kittens and puppies will classify a picture of a kumquat as a kitten or a puppy. An autoencoder watchdog screens trained classifier/regression machine input candidates before processing, e.g. to first test whether the neural network input is a puppy or a kitten. Preliminary results are presented using convolutional neural networks and convolutional autoencoder watchdogs using MNIST images.

Paper Citation

in Harvard Style

Bui J. and Marks II R. (2021). Autoencoder Watchdog Outlier Detection for Classifiers.In Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART, ISBN 978-989-758-484-8, pages 990-996. DOI: 10.5220/0010300509900996

in Bibtex Style

@conference{icaart21,
author={Justin Bui and Robert Marks II},
title={Autoencoder Watchdog Outlier Detection for Classifiers},
booktitle={Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,},
year={2021},
pages={990-996},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010300509900996},
isbn={978-989-758-484-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,
TI - Autoencoder Watchdog Outlier Detection for Classifiers
SN - 978-989-758-484-8
AU - Bui J.
AU - Marks II R.
PY - 2021
SP - 990
EP - 996
DO - 10.5220/0010300509900996