Active sample selection in scalar fields exhibiting non-stationary noise with parametric heteroscedastic Gaussian process regression.
Troy WilsonStefan B. WilliamsPublished in: ICRA (2017)
Keyphrases
- non stationary
- gaussian process regression
- sample selection
- covariance function
- active learning
- gaussian processes
- gaussian process
- noise level
- training data
- missing data
- signal to noise ratio
- change point detection
- support vector machine
- empirical mode decomposition
- gaussian distribution
- computer vision
- noise reduction
- selection strategy
- text classification
- image segmentation