Learning a bi-directional discriminative representation for deep clustering

Wang, Yiming; Chang, Dongxia; Fu, Zhiyuan; Zhao, Yao

doi:10.1016/j.patcog.2022.109237

Cited by 10 publications

(2 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The output is clustered using Gaussian mixture model (GMM), where GMM parameters are updated throughout training. A similar approach was used by Wang et al [11], where they used autoencoders to learn latent representation. Then, they deploy the manifold learning technique UMAP [12] to find a low dimensional space.…”

Section: Graph Clustering Using Deep Networkmentioning

confidence: 99%

Random projection tree similarity metric for SpectralNet

Alshammari¹,

Stavrakakis

Ahmed

et al. 2023

Array

View full text Add to dashboard Cite

Section: Graph Clustering Using Deep Networkmentioning

confidence: 99%

Random projection tree similarity metric for SpectralNet

Alshammari¹,

Stavrakakis

Ahmed

et al. 2023

Array

View full text Add to dashboard Cite

“…In detail, unsupervised representation learning aims to acquire meaningful image features without relying on artificial labels. Among these approaches, clustering-based representation learning methods [6,7] are recognized for their significant potential in this area. However, traditional clustering methods can capture limited intra-image invariance and inter-image similarity.…”

Section: Introductionmentioning

confidence: 99%

Dynamic Graph Clustering Learning for Unsupervised Diabetic Retinopathy Classification

Yu,

Pei

2023

Diagnostics

View full text Add to dashboard Cite

Diabetic retinopathy (DR) is a common complication of diabetes, which can lead to vision loss. Early diagnosis is crucial to prevent the progression of DR. In recent years, deep learning approaches have shown promising results in the development of an intelligent and efficient system for DR classification. However, one major drawback is the need for expert-annotated datasets, which are both time-consuming and costly. To address these challenges, this paper proposes a novel dynamic graph clustering learning (DGCL) method for unsupervised classification of DR, which innovatively deploys the Euclidean and topological features from fundus images for dynamic clustering. Firstly, a multi-structural feature fusion (MFF) module extracts features from the structure of the fundus image and captures topological relationships among multiple samples, generating a fused representation. Secondly, another consistency smoothing clustering (CSC) module combines network updates and deep clustering to ensure stability and smooth performance improvement during model convergence, optimizing the clustering process by iteratively updating the network and refining the clustering results. Lastly, dynamic memory storage is utilized to track and store important information from previous iterations, enhancing the training stability and convergence. During validation, the experimental results with public datasets demonstrated the superiority of our proposed DGCL network.

show abstract