Which Hyperparameters to Optimise? An Investigation of Evolutionary Hyperparameter Optimisation in Graph Neural Network For Molecular Property Prediction.
Yingfang YuanWenjun WangWei PangPublished in: CoRR (2021)
Keyphrases
- hyperparameters
- neural network
- model selection
- cross validation
- bayesian inference
- genetic algorithm
- support vector
- prediction model
- closed form
- random sampling
- gaussian process
- bayesian framework
- prior information
- em algorithm
- noise level
- gaussian processes
- prediction accuracy
- maximum likelihood
- incomplete data
- incremental learning
- maximum a posteriori
- sample size
- regression model
- parameter optimization
- parameter settings
- back propagation
- parameter space
- bp neural network
- grid search
- missing values
- artificial neural networks
- probabilistic model
- active learning
- evolutionary algorithm
- multiscale
- radial basis function
- noise reduction
- error rate
- generative model
- higher order
- image segmentation
- image processing
- feature selection
- machine learning