Autoencoder Watchdog Outlier Detection for Classifiers

TL;DR

This paper introduces an autoencoder watchdog system for classifiers that screens inputs to improve the reliability of neural network predictions, demonstrated with CNNs on MNIST data.

Contribution

It proposes a novel autoencoder-based watchdog mechanism to detect outliers before classification, enhancing neural network robustness.

Findings

Preliminary results show effective outlier detection with convolutional autoencoders.

The approach improves classification reliability on MNIST images.

Autoencoder watchdogs can filter out irrelevant or anomalous inputs.

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 watch dog 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.