Machine Learning and Statistical Learning

Outline

The 24 hours course alternates between lectures and practical sessions.

Week 1

• Introduction

  • Slides

• Optimization problem: gradient descent

  • Slides
  • R Codes

Week 2

• Variants of the gradient descent algorithm

  • Slides
  • R Codes

• Hands-on session: preparing train and test datasets, overfitting

  • Exercise

• Logistic regression

  • Slides
  • Exercise

Week 3

• K-nearest neighbors

  • Slides
  • Exercise

• Support Vector Machines

  • Slides
  • Exercise

• Hands-on: dealing with images

  • jupyter notebook to complete

Week 4

• Working with imbalanced datasets

  • Notebook

• Explainability: Shap Values

  • Slides
  • Notebook

• Machine learning and ethics

  • Slides