IMD1101 Machine Learning

2019.2

• Lesson 01
  • Outline & directions
• Lesson 02
  • Fundamentals of Linear Algebra
  • Introduction to Numpy
  • Boolean Indexing with Numpy
  • Case Study: NYC Taxi-Airport data
• Lesson 03
  • Fundamentals of Machine Learning
  • What is ML?
  • Types
  • Main challenges
  • Test & Validating
• Lesson 04
  • Look at the big picture
  • Get the data
  • Discover and visualize the data to gain insights
  • Prepare the data for Machine Learning algorithms
  • Select a model and train it
  • Fine-tune your model
  • Present your solution
• Lesson 07
  • Univariate and Multivariate KNN
  • Hyperparameter optimization
  • Cross-Validation
  • Pipeline & Gridsearch
• Lesson 09
  • Linear regression
  • Case study: housing price prediction
  • Present the notion of a cost function
  • Introduce the gradient descent method for learning.
  • Refresher on linear algebra concepts.
• Lesson 11
  • First competition
• Lesson 13
  • Loglogistic regression
  • Classification
  • Binary Classification
  • Decision Boundary
  • Cost Function
  • Multiclass Classification
  • Regularization (L1, L2)
  • Hands on Scikit-Learn
• Lesson 17
  • Introduction to Decision Tree
  • Converting categorical variables
  • Splitting Data
  • Decision Trees as flows of data
  • Entropy & Gini
  • Information gain
  • Applying Decision Trees
  • Overfitting problem
  • Case study: classification problem
• Lesson 18
  • Hypothesis test
    • Significant test
    • Chi-squared test
  • Feature Selection for ML
    • Univariate selection
    • Recursive feature elimination
  • Pipelines
• Lesson 19
  • Ensembles (introduction)
  • Voting classifiers
  • Bagging & Pasting
    • Random Forest
    • Feature Importante (XAI)
  • Case Study
• Lesson 20
  • Ensembles (cont.)
  • Boosting
    • Adaboost
    • Gradient Boost
    • XGBoost
• Lesson 21
  • Getting Started with Kaggle
  • Kaggle Workflow
• Lesson 22
  • Clustering Basic
  • K-Means
  • Case study: senators votes, nba
• Lesson 23
  • Representing neural network
  • Nonlinear activation functions
  • Hidden Layers
  • Case study: build a handwritten digit classifier