Methods for choosing a training set which prevents over-learning.
Kazutaka YamasakiHidemitsu OgawaPublished in: Systems and Computers in Japan (1994)
Keyphrases
- training set
- supervised learning
- empirical studies
- significant improvement
- cross validation
- preprocessing
- learning systems
- learning algorithm
- learning models
- human experts
- benchmark datasets
- reinforcement learning
- computational cost
- prior knowledge
- learning process
- qualitative and quantitative
- active learning
- online learning
- knowledge acquisition
- pairwise
- support vector
- learning tasks
- data sets
- bayesian methods
- training set size