Reliable adaptive distributed hyperparameter optimization (RadHPO) for deep learning training and uncertainty estimation.
John LiMaria PantojaGerardo Fernández-EscribanoPublished in: J. Supercomput. (2023)
Keyphrases
- computer vision
- deep learning
- deep architectures
- restricted boltzmann machine
- machine learning
- unsupervised learning
- unsupervised feature learning
- mental models
- deep belief networks
- weakly supervised
- supervised learning
- online learning
- model selection
- data sets
- viewpoint
- pairwise
- decision trees
- training examples
- learning strategies