Fundamentals of Machine Learning

A series of 8 lectures for "The Machine Learning for Advanced Integrated Intelligence alpha" course
Taught by Emtiyaz Khan, AIP, RIKEN
E-mail at emtiyaz.khan@riken.jp
Goals Lecture-Notes

News

Information about the exam to be held on May 12, 2018

  • Update on May 12 at 1pm JST. Sample answers available.
  • Here is a sample exam.
  • The exam starts at 16H00 and ends at 17H30 (90 minutes in total).
  • No documents allowed apart from one A4 sheet of your own notes.
  • No electronic devices are allowed except a calculator. Make sure that your calculator is only a calculator and cannot be used for any other purpose other than for manual numerical-calculations.

Information for the Lecture on April 29, 2018

  • Part 2 of lecture notes are posted now.
  • I have updated the annotated notes.

Information for the Lecture on April 21, 2018

  • I have posted the annotated version after the lecture on April 21, 2018.
  • You can ask me questions over Twitter (@EmtiyazKhan).
  • To do for this week:
    • learn about linear algebra,
    • learn about taking derivative of the MSE cost function,
    • learn about computational complexity,
    • review the course material,
    • try the lab 1 and 2,
    • read the material for the next lecture,
    • submit one page report before the next lecture.

Information for the Lecture on April 14, 2018

  • I have posted the annotated version after the lecture on April 14, 2018.
  • To do for this week: learn about linear algebra, review course material, try the lab 1 and 2, read the material for the next lecture, submit one page report before the next lecture.

Information for the Lecture on April 14, 2018

  • Lecture notes are available below.
  • We will provide a printed copy of the lecture notes before the lecture.
  • We will also give you a blank page. Please write your student ID and name on top of this page. During the lecture, we will have several quizzes. Please write down your answers in this sheet. We will also ask you to write down your own summary of the lecture. Please write those in this sheet. At the end of the lecture, don't forget to submit this sheet. This will be part of the course evaluations (30% of the final marks).
  • You need to know how to multiply matrices and vectors. Here is a link to learn about it.
  • You need to know how to take derivatives with respect to vectors. Here is a link, and another one here.
  • A short tutorial which summarizes all linear algebra you need to know for this course.

Lecture Notes

Introduction to Machine Learning Slides
Course Information Slides
Linear Algebra (Handout) Slides
Part 1 (Linear Regression, SGD, LeastSquares, Overfitting, CV, Bias-Variance) Notes Annotated Notes
Part 2 (Classification with k-NN, Logistic Regression, Neural Networks) Notes

Lab Exercises

You are encouraged to try these exercises. These may help you learn faster and better. If you try them, write about it in your weekly report. Note that these are not mandatory to do.
  1. Basic Python and Matlab
  2. Linear Regression and Gradient Descent
  3. Least Squares and Overfitting
  4. CrossValidation and Overfitting

Course Goals

Evaluation is based on the following methods:
  • 30% Weekly reports
  • 30% Class participation
  • 40% Exams
By the end of the course, the student must be able to:
  • Define Regression and Classification, and explain the main differences between them
  • Describe a few models and algorithms for them.
  • Implement and apply these methods.
  • Derive the theory behind ML methods taught in the course and generalize them to new problems.
  • Continue to work through difficulties or initial failure to find optimal solutions.
  • Assess one’s own level of skill acquisition, and plan their on-going learning goals.