Recursive quality optimization of a smart forming tool under the use of perception based hybrid datasets for training of a Deep Neural Network.
Sebastian FeldmannMichael SchmiedtJulian Marc SchlosserWolfgang RimkusTobias StempfleChristian RathmannPublished in: Discov. Artif. Intell. (2022)
Keyphrases
- neural network
- training process
- feed forward neural networks
- training algorithm
- feedforward neural networks
- multilayer neural network
- multi layer perceptron
- training dataset
- optimization problems
- benchmark datasets
- recurrent networks
- global optimization
- neural network model
- image reconstruction from projections
- training patterns
- optimization process
- artificial neural networks
- human perception
- database
- training data
- deep architectures
- high quality
- stochastic gradient descent
- training samples
- fault diagnosis
- training examples
- test set
- fuzzy logic
- network architecture
- constrained optimization
- neural network training
- optimization method
- training and testing data
- data quality
- hybrid optimization algorithm
- quality measures