DKM: Differentiable K-Means Clustering Layer for Neural Network Compression.
Minsik ChoKeivan Alizadeh-VahidSaurabh AdyaMohammad RastegariPublished in: CoRR (2021)
Keyphrases
- neural network
- multi layer
- data compression
- image compression
- genetic algorithm
- single layer
- objective function
- compression scheme
- neural network model
- compression ratio
- artificial neural networks
- pattern recognition
- loss function
- fault diagnosis
- self organizing maps
- clustering algorithm
- feed forward neural networks
- back propagation
- compression algorithm
- neural nets
- fuzzy logic
- feed forward
- spectral clustering
- radial basis function
- control system
- training algorithm
- compression rate
- backpropagation neural network
- neural network is trained
- fermentation process
- image reconstruction from projections