

Introduction to Machine Learning

Note

The content of this website was automatically generated from Jupyter notebooks.

Hopefully, nothing has been broken.

Lectures notes are prepared in Colaboratory, which is a Jupyter notebook environment running in the cloud and storing notebooks in Google Drive. It makes it easy to collaborate on a project in Jupyter notebooks and provides free computing power.

This website was created for your convenience. Feel free to use Jupyter notebooks though:

• Introduction (Colaboratory or GitHub)

• k-Nearest Neighbors (Colaboratory or GitHub)

  • First ML problem

  • Nearest Neighbor algorithm

  • k-Nearest Neighbors algorithm

  • Hyperparameters

  • Iris dataset

  • MNIST

  • Regression with kNN

  • Summary

• Decision trees (Colaboratory or GitHub)

  • Introduction

  • ID3 and C4.5 algorithms

  • CART

  • Bias-variance trade-off

  • Ensemble learning

  • Summary

• Support vector machine (Colaboratory or GitHub)

  • Linear SVM

  • Lagrange multipliers

  • Optimal margin

  • Non-linear SVM

  • Soft margin

  • SMO algorithm

  • Examples

  • Multiclass classification

  • SVM regression

  • Summary

• Neural networks (Colaboratory or GitHub)

  • Linear regression

  • Logistic regression

  • Multinominal logistic regression

  • Neural networks

  • AND, OR, XOR

  • Simple regression with NN

  • MNIST with softmax

  • Gradient descent variations

  • Regularization

  • Summary

• Deep learning (Colaboratory or GitHub)

  • Convolutional Neural Networks

  • Deep MNIST

  • Batch normalization

  • Data augmentation

  • Summary