Learning Robust Representation for Clustering through Locality Preserving Variational Discriminative Network
This work provides an incremental improvement to the VaDE clustering method, making it more robust to noise and better at preserving local structures, which is beneficial for practitioners using deep clustering algorithms.
This paper addresses the fragility of Variational Deep Embedding (VaDE) to input noise and its neglect of locality information in clustering. The authors propose a joint learning framework that enhances VaDE with a robust embedding discriminator and a local structure constraint, leading to state-of-the-art performance across various vision and textual datasets and improved robustness to adversarial inputs.
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.