Structure-Driven Representation Learning for Deep Clustering
Open Access
- 16 October 2023
- journal article
- research article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Knowledge Discovery From Data
- Vol. 18 (1) , 1-25
Abstract
As an important branch of unsupervised learning methods, clustering makes a wide contribution in the area of data mining. It is well known that capturing the group-discriminative properties of each sample for clustering is crucial. Among them, deep clustering delivers promising results due to the strong representational power of neural networks. However, most of them adopt sample-level learning strategies, and the standalone data point barely captures its holistic cluster’s context and may undergo sub-optimal cluster assignment. To tackle this issue, we propose a Structure-driven Representation Learning (SRL) method by introducing latent structure information into the representation learning process at both the local and global levels. Specifically, a local-structure-driven sample representation strategy is proposed to approximate the estimation of data distribution, which models the neighborhood distribution of samples with potential structure information and exploits statistical dependencies between them to improve cluster consistency. A global-structure-driven cluster representation strategy is designed, where the context of each cluster is sufficiently encoded according to its samples (exemplar-theory) and corresponding prototype (prototype-theory). In this case, each cluster can only be related to its most similar samples, and different clusters are separated as much as possible. These two models are seamlessly combined into a joint optimization problem, which can be efficiently solved. Experiments on six widely-used datasets demonstrate the superiority of SRL over state-of-the-art clustering methods.Keywords
Funding Information
- National Natural Science Foundation of China (62176020)
- National key research and development program (2020AAA0106800)
- Joint Foundation of the Ministry of Education (8091B042235)
- Beijing Natural Science Foundation (L211016)
- Fundamental Research Funds for the Central Universities (2019JBZ110)
- Chinese Academy of Sciences (OEIP-O-202004)
This publication has 34 references indexed in Scilit:
- Dynamic Affinity Graph Construction for Spectral Clustering Using Multiple FeaturesIEEE Transactions on Neural Networks and Learning Systems, 2018
- Rank-Constrained Spectral Clustering With Flexible EmbeddingIEEE Transactions on Neural Networks and Learning Systems, 2018
- A review of clustering techniques and developmentsNeurocomputing, 2017
- Deep Adaptive Image Clustering2017
- Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization2017
- Variational Deep Embedding: An Unsupervised and Generative Approach to Clustering2017
- Joint Unsupervised Learning of Deep Representations and Image Clusters2016
- Deep Residual Learning for Image Recognition2016
- Data clustering: 50 years beyond K-meansPattern Recognition Letters, 2010
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998