Basics of Machine Learning

The following elementary concepts of machine learning are introduced:

  • supervised training with data
    • structure of training data
    • (linear) regression
    • classification (logistic regression)
  • maximum likelihood principle
  • gradient descent
  • cost function and loss
    • squared error loss
    • cross entropy loss
  • linear separability, non-linear transformations with basis functions
  • overfitting, underfitting, bias-variance tradeoff
  • learning curves
  • regularization
  • validation