NAS-HPO-Bench-II: A Benchmark Dataset on Joint Optimization of Convolutional Neural Network Architecture and Training Hyperparameters.
Yoichi HiroseNozomu YoshinariShinichi ShirakawaPublished in: ACML (2021)
Keyphrases
- benchmark datasets
- joint optimization
- hyperparameters
- model selection
- cross validation
- support vector
- bayesian inference
- random sampling
- closed form
- grid search
- em algorithm
- gaussian process
- noise level
- bayesian framework
- maximum likelihood
- sparse representation
- sample size
- prior information
- training set
- data sets
- incomplete data
- maximum a posteriori
- incremental learning
- pattern recognition
- probabilistic model
- parameter space
- parameter settings
- active learning
- generalization ability
- training data
- feature extraction
- support vector machine
- neural network
- principal component analysis