1 code implementation • ICLR 2021 • Yaling Tao, Kentaro Takagi, Kouta Nakata
In this paper, we propose a clustering-friendly representation learning method using instance discrimination and feature decorrelation.
Ranked #6 on Image Clustering on ImageNet-10
no code implementations • 6 Dec 2018 • Yaling Tao, Kentaro Takagi, Kouta Nakata
DEC optimizes cluster assignments by pushing data more densely around centroids in latent space, but it is sometimes sensitive to the initial location of centroids, especially in the case of imbalanced data, where the minor class has less chance to be assigned a good centroid.