Component-Wise Natural Gradient Descent - An Efficient Neural Network Optimization.
Tran Van SangMhd IrvanRie Shigetomi YamaguchiToshiyuki NakataPublished in: CANDAR (2022)
Keyphrases
- neural network
- cost function
- artificial neural networks
- back propagation
- highly non linear
- optimization process
- optimization problems
- optimization algorithm
- image reconstruction from projections
- optimization model
- global optimization
- optimization method
- self organizing maps
- fault diagnosis
- real world
- knn
- multi objective
- training data
- objective function
- neural nets
- genetic algorithm
- auto associative
- stochastic gradient
- feed forward neural networks
- constrained optimization
- multi layer
- computationally efficient
- bp neural network
- neural network model
- pattern recognition
- pairwise
- fuzzy logic
- active contours