Hyper-Parameter Selection in Convolutional Neural Networks Using Microcanonical Optimization Algorithm.
Ayla GülcüZeki KusPublished in: IEEE Access (2020)
Keyphrases
- optimization algorithm
- convolutional neural networks
- hyperparameters
- multi objective
- optimization method
- particle swarm optimization pso
- cross validation
- differential evolution
- parameter settings
- model selection
- convolutional network
- evolutionary multi objective
- optimization strategy
- hybrid optimization algorithm
- closed form
- feature selection
- artificial bee colony
- bayesian inference
- neural network
- training data
- objective function
- random sampling
- em algorithm
- search space
- bayesian framework
- learning process
- evolutionary algorithm
- np hard
- super resolution
- maximum likelihood