Supervised learning and regression

Prediction from labelled data: least squares, ridge, lasso and kernel or Gaussian-process regression (including GBLUP), the bias-variance decomposition, Bayes classifiers and optimal predictors, log loss, decision-tree information gain, backpropagation, and sufficient input subsets.

• Backpropagation
• Bayes classifier
• Bayes estimators for squared and absolute loss
• Bias–variance decomposition
• Cross-entropy loss
• Gaussian process regression
• Information gain in decision trees
• Least squares, ridge and lasso regression
• Kernel ridge regression and GBLUP
• Sufficient input subsets

The dependency explorer needs JavaScript and its graph data.

Download the complete graph data.