Consistency and adaptivity are complementary targets for the validation of variance-based uncertainty quantification metrics in machine learning regression tasks.
Pascal PernotPublished in: CoRR (2023)
Keyphrases
- machine learning
- model selection
- machine learning algorithms
- regression model
- transfer learning
- learning tasks
- classification and regression problems
- data mining
- explanation based learning
- computer science
- artificial intelligence
- information extraction
- feature selection
- pattern recognition
- multi task
- regression analysis
- machine learning approaches
- inductive logic programming
- machine learning methods
- decision trees
- support vector
- decision theory
- knowledge acquisition
- text classification
- feature engineering
- least squares
- target detection
- knowledge discovery
- learning algorithm
- evaluation metrics
- linear regression
- standard deviation
- covariance matrix
- cross validation
- learning systems
- computational intelligence
- text mining