Supervised learning

In this passage you will have a chance to work with Polynomial, Linear and Logistic regressions, Decision Trees and Support Vector Machines. The section is divided into two subcategories : regression and classification. We strongly recommend you to start with regression category, as main concepts of machine learning are exposed there.

Problems:

• Regression
  • Linear regression and core concepts of machine learning
    • Basics of linear regression and loss functions
    • Transferring and processing of data
    • Train/Test split and model validation
    • Description of assignment
  • Polynomial regression
    • Key concepts of polynomial regression
    • Description of assignment
  • l1 and l2 regularization
    • Overfitting problem
    • Underfitting problem
    • Regularization
    • Feature normalization
    • Description of assignment
  • SVM (Support Vector Machines)
    • The gist of SVM
    • Handling Overfitting
    • Description of assignment
  • Decision trees family
    • Decision trees
    • Random forest
    • GBDT (Gradient Boosted Decision Trees)
    • Description of assignment
• Classification
  • Binary classification and core concepts of it
    • Differences between classification and regression
    • Classification metrics
    • Cross validation and greedy search
    • Description of assignment
  • Hyperparameters tuning
    • Grid search
    • Randomized parameter optimization
    • Tuning through minimization
    • Description of assignment
  • Multiclassification
    • Description of assignment