top of page

Machine Learning:

Classification

Week 1

Linear Classifiers & Logistic Regression

(080616)

  • Linear classifiers
  • Class probabilities
  • Logistic regression
  • Practical issues for classification
  • Summarizing linear classifiers & logistic regression
  • Programming Assignment
  • Quiz: Linear Classifiers & Logistic Regression
  • Quiz: Predicting sentiment from product reviews
Week 2

Learning Linear Classifiers

Overfitting & Regularization in Logistic Regression

(080616)

  • Maximum likelihood estimation

  • Gradient ascent algorithm for learning logistic regression classifier

  • Choosing step size for gradient ascent/descent

  • (VERY OPTIONAL LESSON) Deriving gradient of logistic regression

  • Summarizing learning linear classifiers

  • Programming Assignment

  • Overfitting in classification

  • Overconfident predictions due to overfitting

Week 2

Learning Linear Classifiers

Overfitting & Regularization in Logistic Regression

(080616)

  • L2 regularized logistic regression

  • Sparse logistic regression

  • Summarizing overfitting & regularization in logistic regression

  • Programming Assignment

  • Quiz: Learning Linear Classifiers

  • Quiz: Implementing logistic regression from scratch

  • Quiz: Overfitting & Regularization in Logistic Regression

  • Quiz: Logistic Regression with L2 regularization

Week 3

Decision Trees

(080616)

  • Intuition behind decision trees

  • Learning decision trees

  • Using the learned decision tree

  • Learning decision trees with continuous inputs

  • Summarizing decision trees

  • Programming Assignment 1

  • Programming Assignment 2

  • Quiz: Decision Trees

  • Quiz: Identifying safe loans with decision trees

  • Quiz: Implementing binary decision trees

Week 4

Preventing Overfitting in Decision Trees

Handling Missing Data

(080616)

  • Overfitting in decision trees

  • Early stopping to avoid overfitting

  • (OPTIONAL LESSON) Pruning decision trees

  • Summarizing preventing overfitting in decision trees

  • Programming Assignment

  • Basic strategies for handling missing data

Week 4

Preventing Overfitting in Decision Trees

Handling Missing Data

(080616)

  • Strategy 3: Modify learning algorithm to explicitly handle missing data

  • Summarizing handling missing data

  • Quiz: Preventing Overfitting in Decision Trees

  • Quiz: Decision Trees in Practice

  • Quiz: Handling Missing Data

Week 5

Boosting

(080616)

  • The amazing idea of boosting a classifier

  • AdaBoost

  • Applying AdaBoost

  • Programming Assignment 1

  • Convergence and overfitting in boosting

  • Summarizing boosting

  • Programming Assignment 2

  • Quiz: Exploring Ensemble Methods

  • Quiz: Boosting

  • Quiz: Boosting a decision stump

Week 6

Precision-Recall

(080616)

  • Why use precision & recall as quality metrics

  • Precision & recall explained

  • The precision-recall tradeoff

  • Summarizing precision-recall

  • Programming Assignment

  • Quiz: Precision-Recall

  • Quiz: Exploring precision and recall

Week 7

Scaling to Huge Datasets &

Online Learning

(080616)

  • Scaling ML to huge datasets

  • Scaling ML with stochastic gradient

  • Understanding why stochastic gradient works

  • Stochastic gradient: Practical tricks

  • Online learning: Fitting models from streaming data

  • Summarizing scaling to huge datasets & online learning

  • Programming Assignment

  • Quiz: Scaling to Huge Datasets & Online Learning

  • Quiz: Training Logistic Regression via Stochastic Gradient Ascent

bottom of page