The Labeling Distribution Matrix (LDM): A Tool for Estimating Machine Learning Algorithm Capacity.
Pedro Sandoval SeguraJulius LauwDaniel BashirKinjal ShahSonia SehraDominique MaciasGeorge D. MontañezPublished in: ICAART (2) (2020)
Keyphrases
- learning algorithm
- active learning
- training data
- singular value decomposition
- probability distribution
- image segmentation
- learning scheme
- rigid body
- gaussian distribution
- labeling process
- laplace transform
- reinforcement learning
- unsupervised learning
- labeled data
- machine learning algorithms
- neural network
- generalization ability
- software tools
- spatial distribution
- power law