Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Deep Discriminative Clustering Analysis

About

Traditional clustering methods often perform clustering with low-level indiscriminative representations and ignore relationships between patterns, resulting in slight achievements in the era of deep learning. To handle this problem, we develop Deep Discriminative Clustering (DDC) that models the clustering task by investigating relationships between patterns with a deep neural network. Technically, a global constraint is introduced to adaptively estimate the relationships, and a local constraint is developed to endow the network with the capability of learning high-level discriminative representations. By iteratively training the network and estimating the relationships in a mini-batch manner, DDC theoretically converges and the trained network enables to generate a group of discriminative representations that can be treated as clustering centers for straightway clustering. Extensive experiments strongly demonstrate that DDC outperforms current methods on eight image, text and audio datasets concurrently.

Jianlong Chang, Yiwen Guo, Lingfeng Wang, Gaofeng Meng, Shiming Xiang, Chunhong Pan• 2019

Related benchmarks

TaskDatasetResultRank
Image ClusteringCIFAR-10
NMI0.424
243
Image ClusteringSTL-10
ACC48.9
229
ClusteringCIFAR-10 (test)
Accuracy52.4
184
Image ClusteringImageNet-10
NMI0.433
166
ClusteringSTL-10 (test)
Accuracy48.9
146
ClusteringImageNet-10 (test)
ACC57.7
69
Image ClusteringCIFAR-10 (full)
Accuracy0.524
33
Showing 7 of 7 rows

Other info

Follow for update