Pattern Classification and Machine Learning

Taught by Emtiyaz Khan, EPFL (IC)
E-mail at pcml.epfl [at] (no personal emails please!)
Goals Lecture-Notes Project-1 Project-2 TAs News

Mock midterm & solutions

The mock midterm and the corresponding solutions are available.

PCML 2014 course summary published

The course summary from last year's iteration of the class was released. It is posted as-in, and may contain a mismatches with this year's iteration.

Project I stats

Stats for Project I grades have been posted.

Practice questions

Some exercise questions that will help you understand the text better. Remember, these DO NOT represent the questions that you will see in the exam!

Course Timings

  • Lecture (8h15-10h) on Tuesdays in CE-1
  • Lecture (8h15-10h) on Thursdays in CE-4
  • Lab (14h15-16h) on Thursday in INJ-218

Lecture Notes

Disclaimer: There might be technical and grammatical mistakes in these notes since I do not have time to write carefully. My apologies. Please use at your own risk. Please email me if you see any mistakes.
PCML 2014 course summary (not updated) PDF
Lab: Gaussian Processes Sheet Code and Data
Random Forests Notes
Decision Trees Notes
Gaussian Processes Notes Annotated
Q&A for Project II, feedback on Project I
Multi-Layer Perceptron Notes Annotated
BayesNet and Belief Propagation Notes Annotated
Lab: SVM Sheet Code and Data
Examples of Time-series Notes
SVD and PCA Notes Annotated
Lab: Recommendation systems Sheet Code and Data
Matrix factorization Notes Annotated
Mock exam Sheet Solutions
EM algorithm Notes Annotated
Gaussian Mixture Model Notes Annotated
Lab: K-means Sheet Code
K-means Notes Annotated
Unsupervised Learning Notes A note by P. Dayan
Lab: Q&A on Project-I
Support Vector Machines Notes Annotated
Kernel Ridge Regression Notes Annotated, A note by M. Seeger
Lab: Q&A on Project-I
Curse of dimensionality and kNN Notes Annotated
Generalized Linear Model Notes Annotated
Lab: Logistic regression Sheet
Logistic regression Notes Annotated
Classification Notes Annotated
Lab: Cross validation & bias-variance decomposition Sheet cvDemo.m
Bias-Variance decomposition Notes Annotated
Cross-Validation Notes Annotated
Lab: Testing Linear Regression Sheet Code & data
Ridge Regression Notes Annotated
Overfitting Notes Annotated
Maximum Likelihood Notes Annotated
Least Squares Notes Annotated, PDF on ill-conditioning
Lab: Linear Regression & Gradient Descent Sheet gradientDescent.m, gridSearch.m
Gradient Descent Notes Annotations
Cost Functions Notes Annotations
Lab: Introduction to Matlab Sheet Data
Linear Regression Notes Annotations
Regression Notes Annotations
Linear Algebra Handout Handout 2
Course Information Notes
Introduction to PCML Slides Teacher's Slides

Course Goals

Evaluation is based on the following methods.
  • 10% Project-I
  • 30% Project-II
  • 60% Exams
By the end of the course, the student must be able to:
  • Define the following basic machine learning problems and explain main differences between them: Regression, classification, clustering, dimensionality reduction, time-series.
  • Describe a few important models and algorithms for the basic ML problems.
  • Implement, apply, and compare these methods to real-world problems.
  • Choose a method for the real-world problem in hand.
  • Critique and defend your choice of method.
  • Derive the theory behind ML methods taught in the course and generalize them to new problems.

Teaching Assistants

  • Becker Carlos Joaquin
  • Bermudez Chacon Roger
  • Newton Taylor Howard
  • Salehi Farnood
  • Seguin Benoit Laurent Auguste
  • Victor Kristof
  • Dennis Meier
  • Merlin Nimier-David
  • Jakub Sygnowski (voluntary TA)
  • Michalina Pacholska (voluntary TA)