Learning Robust Representation for Clustering through Locality Preserving Variational Discriminative Network

TL;DR

This paper introduces a robust clustering method that enhances variational deep embedding by incorporating local structure constraints and a discriminator, leading to improved robustness and performance on vision and textual datasets.

Contribution

It proposes a joint learning framework that addresses VaDE's fragility to noise and lack of locality information by adding a local structure constraint and a robust embedding discriminator.

Findings

Outperforms state-of-the-art models on multiple datasets

Demonstrates high robustness to adversarial inputs

Improves clustering accuracy and stability

Abstract

Clustering is one of the fundamental problems in unsupervised learning. Recent deep learning based methods focus on learning clustering oriented representations. Among those methods, Variational Deep Embedding achieves great success in various clustering tasks by specifying a Gaussian Mixture prior to the latent space. However, VaDE suffers from two problems: 1) it is fragile to the input noise; 2) it ignores the locality information between the neighboring data points. In this paper, we propose a joint learning framework that improves VaDE with a robust embedding discriminator and a local structure constraint, which are both helpful to improve the robustness of our model. Experiment results on various vision and textual datasets demonstrate that our method outperforms the state-of-the-art baseline models in all metrics. Further detailed analysis shows that our proposed model is very robust to the adversarial inputs, which is a desirable property for practical applications.