# Autoencoders as Pattern Filters

## TL;DR

This paper presents a straightforward method to repurpose autoencoders as pattern filters and robust classifiers by training them to selectively filter specific class patterns, enhancing their filtering and classification capabilities.

## Contribution

The paper introduces a simple technique to transform autoencoders into pattern filters and classifiers, enabling targeted pattern filtering and improved robustness.

## Key findings

- Autoencoders can be effectively used as pattern filters.
- The approach improves classification robustness by filtering specific class patterns.
- The method is simple and adaptable for various pattern recognition tasks.

## Abstract

We discuss a simple approach to transform autoencoders into "pattern filters". Besides filtering, we show how this simple approach can be used also to build robust classifiers, by learning to filter only patterns of a given class.