Efficiently Computable Safety Bounds for Gaussian Processes in Active Learning.
Jörn TebbeChristoph ZimmerAnsgar StelandMarkus Lange-HegermannFabian MiesPublished in: CoRR (2024)
Keyphrases
- gaussian processes
- efficiently computable
- active learning
- upper bound
- sufficient conditions
- gaussian process
- lower bound
- computationally hard
- random sampling
- hyperparameters
- semi supervised
- gaussian process regression
- covariance function
- learning algorithm
- gaussian process models
- winnow algorithm
- vc dimension
- worst case
- training set
- supervised learning
- machine learning
- sample complexity
- perceptron algorithm
- data sets
- generalization error
- multi task