# Deep Spectral Clustering using Dual Autoencoder Network

## TL;DR

This paper introduces a deep spectral clustering framework that uses a dual autoencoder network to learn robust, discriminative embeddings for improved clustering performance, outperforming existing methods on benchmark datasets.

## Contribution

It presents a novel joint learning framework combining a dual autoencoder and spectral clustering, enhancing robustness and discriminability of embeddings for clustering tasks.

## Key findings

- Significantly outperforms state-of-the-art clustering methods on benchmark datasets.
- The dual autoencoder improves robustness of latent representations to noise.
- Spectral clustering on learned embeddings yields more accurate clustering results.

## Abstract

The clustering methods have recently absorbed even-increasing attention in learning and vision. Deep clustering combines embedding and clustering together to obtain optimal embedding subspace for clustering, which can be more effective compared with conventional clustering methods. In this paper, we propose a joint learning framework for discriminative embedding and spectral clustering. We first devise a dual autoencoder network, which enforces the reconstruction constraint for the latent representations and their noisy versions, to embed the inputs into a latent space for clustering. As such the learned latent representations can be more robust to noise. Then the mutual information estimation is utilized to provide more discriminative information from the inputs. Furthermore, a deep spectral clustering method is applied to embed the latent representations into the eigenspace and subsequently clusters them, which can fully exploit the relationship between inputs to achieve optimal clustering results. Experimental results on benchmark datasets show that our method can significantly outperform state-of-the-art clustering approaches.