A Two-Stage Subspace Trust Region Approach for Deep Neural Network Training.
Viacheslav DudarGiovanni ChierchiaÉmilie ChouzenouxJean-Christophe PesquetVladimir SemenovPublished in: CoRR (2018)
Keyphrases
- trust region
- neural network training
- optimization method
- global optimum
- training algorithm
- optimization methods
- neural network
- levenberg marquardt
- column generation
- newton method
- hessian matrix
- line search
- log likelihood
- back propagation
- simulated annealing
- principal component analysis
- optimization algorithm
- particle swarm optimisation
- feature extraction
- data sets
- genetic algorithm
- optimization problems
- cost function
- multi objective
- search algorithm
- step size
- clustering algorithm
- linear equations
- faster convergence
- global convergence
- pairwise
- branch and bound
- search space
- differential evolution