Deep k-Means: Re-Training and Parameter Sharing with Harder Cluster Assignments for Compressing Deep Convolutions.
Junru WuYue WangZhenyu WuZhangyang WangAshok VeeraraghavanYingyan LinPublished in: ICML (2018)
Keyphrases
- k means
- clustering algorithm
- deep architectures
- cluster analysis
- deep learning
- data clustering
- clustering framework
- unsupervised clustering
- information sharing
- hierarchical clustering
- data sets
- clustering method
- np complete
- data compression
- cluster centers
- constrained clustering
- data points
- hierarchical clustering algorithm
- agglomerative hierarchical clustering
- active learning
- variable weighting