Automating Configuration of Convolutional Neural Network Hyperparameters Using Genetic Algorithm.
Franklin JohnsonAlvaro ValderramaCarlos ValleBroderick CrawfordRicardo SotoRicardo ÑanculefPublished in: IEEE Access (2020)
Keyphrases
- hyperparameters
- convolutional neural network
- genetic algorithm
- model selection
- cross validation
- bayesian inference
- closed form
- random sampling
- bayesian framework
- face detection
- gaussian process
- prior information
- support vector
- sample size
- em algorithm
- noise level
- maximum likelihood
- parameter optimization
- gaussian processes
- incremental learning
- neural network
- maximum a posteriori
- missing values
- incomplete data
- evolutionary algorithm
- parameter settings
- parameter space
- data sets
- fitness function
- simulated annealing
- object recognition
- machine learning
- feature selection
- genetic programming
- object detection
- computer vision