# Deep Clustering Based on a Mixture of Autoencoders

## TL;DR

This paper introduces DAMIC, a deep clustering algorithm using a mixture of autoencoders where each cluster is modeled by an autoencoder, jointly learning data representation and cluster assignment without regularization.

## Contribution

The paper presents a novel deep autoencoder mixture model for clustering that avoids data collapse and improves performance over existing methods.

## Key findings

- Significant improvement over state-of-the-art methods on image and text data
- No regularization needed to prevent data collapse
- Joint learning of data representation and clustering enhances accuracy

## Abstract

In this paper we propose a Deep Autoencoder MIxture Clustering (DAMIC) algorithm based on a mixture of deep autoencoders where each cluster is represented by an autoencoder. A clustering network transforms the data into another space and then selects one of the clusters. Next, the autoencoder associated with this cluster is used to reconstruct the data-point. The clustering algorithm jointly learns the nonlinear data representation and the set of autoencoders. The optimal clustering is found by minimizing the reconstruction loss of the mixture of autoencoder network. Unlike other deep clustering algorithms, no regularization term is needed to avoid data collapsing to a single point. Our experimental evaluations on image and text corpora show significant improvement over state-of-the-art methods.