Post-training deep neural network pruning via layer-wise calibration.
Ivan LazarevichAlexander KozlovNikita MalininPublished in: ICCVW (2021)
Keyphrases
- neural network
- feed forward neural networks
- training algorithm
- multi layer
- training process
- feedforward neural networks
- multi layer perceptron
- single layer
- neural network training
- back propagation
- multiple layers
- backpropagation algorithm
- camera calibration
- neural network model
- output layer
- pruning method
- error back propagation
- training phase
- avoid overfitting
- train a neural network
- genetic algorithm
- pattern recognition
- artificial neural networks
- online learning
- neural nets
- training patterns
- deep learning
- restricted boltzmann machine
- test set
- feed forward
- hidden layer
- data sets
- fuzzy logic
- training examples
- knn
- recurrent networks
- pairwise
- pruning methods
- training data