Home
Course Guidelines
About the course Prerequite Material References Extra Material Errata to Video Welcome
Python
Jupyter Notebooks Python overview
Exercises
Before semester start: Installation Week 1: Introduction to Python and libraries Week 2: Vector representations Week 3: Linear Algebra Week 4: Linear Transformations Week 5: Models and least squares Week 6: Assignment 1 - Gaze Estimation Week 7: Model selection and descriptive statistics Week 8: Filtering Week 9: Classification Week 10: Evaluation Week 11: Dimensionality reduction Week 12: Clustering and refresh on gradients Week 13: Neural Networks Week 14: Convolutional Neural Networks (CNN's)
Tutorials
Week 2: Data analysis, manipulation and plotting Week 3: Linear algebra Week 4: Transformations tutorial Week 5: Projection and Least Squares tutorial Week 7: Cross-validation and descriptive statistics tutorial Week 8: Filtering tutorial Week 11: Gradient Descent / Ascent
In-class Exercises
In-class 1 In-class 2 In-class 10 In-class 3 In-class 4 In-class 8

Document

  • Overview
  • 1. Material Treasurehunt
  • 2. Practical introduction to Python and Numpy

Content

  • Task 1 Course Overview
  • Task 2 Course Overview
  • Task 3 Preparation
  • Task 4 Learning activities
  • Task 5 Excercises
  • Task 6 Mandatory
  • Task 7 Exams

%var|reveal_date("2025-05-24")

Guided tour of the course material

This exercise is about navigating through and being able to locate the course material. Initially, the TAs will provide a guided tour of the material, after which you will be asked to locate specific treasures (pieces of information) within the course material. The purpose of the questions is for you to become familiar with the structure of the materials. These exercises can be completed outside the exercise classes and you do not need to do them all (see icons of the exercise). Concentrate on the ones you find most useful for your understanding and ask the TAs or Dan in case you are in doubt. Some of the answers will be released later.

Live walk through

TAs will give a live overview of the course material structure, including:

  • Lecture preparation and exercises on LearnIT
  • The IML webpage
  • Grasple
  • OneDrive (location of exercises)
  • test
Task 1: Course Overview
  1. Where is the exercise material for week 1 (jupyter notebooks) located
  2. Who is the author of [SH]
  3. What does the picture of [PM] page 20 contain?
  4. Who is the teacher in the video ("Linear transformations and their matrices") of week 3 in the extra material.
  5. Are the exercises and assignments part of the exam?
  6. How many different focus points were there last year for exam question 3 and 6.
Answers and suggestions:

Only few answers are provided here beside 5. Yes, especially the exercises marked as "expected" (see icons)

Ask if you are in doubt about the others

Task 2: Course Overview
  1. Which programming and mathematical topics are you expected to be familiar with and able to use in practice before the course?
  2. List the icons used in the exercises and explain what each one represents.
  3. Explain the meaning of the icons in the exercises on this page an in the next exercise .
Answers and suggestions:
  1. See iml.itu.dk under "Course Guidelines / Prerequisite Material".
  2. The icons appear on the front page of most exercise weeks e.g., https://iml.itu.dk/03-exercises/W01/index.html
Reflections on introduction video and learning activities

The following questions relate to the introduction video and the main components of the course.

Task 4: Learning activities
  1. Use the official course description, introduction video, and course guidelines here on iml.itu.dk to determine which learning activities are part of the course.
  2. List the topics covered in the course.
  3. Why is Part I of the course essential for the course?
  4. What are the four main categories of machine learning models introduced in the course?
  5. Where can you find which parts of the course have changed compared to what is mentioned in the video?
  6. How many people must be in each group, and by when must groups be formed?
  7. How many hours of work are you expected to do on average each week for the course?
Answers and suggestions:
  1. The introduction video (around 8 minutes) for an overview of the learning activities, including lectures, exercises, in-class exercises, tutorials, and assignments.

  2. Refer to LearnIT for the weekly topics and titles. Each week may also include a short description of the content.

  3. Most of the mathematical notation and conceptual foundation needed for to machine learning is established in Part I (video around 8.30). The subsequent parts of the course build heavily on this material. It is therefore essential that you attend the lectures, complete as many exercises as possible, and ask questions whenever something is unclear. Even if you only grasp 30% at first, it is important to address uncertainties early to establish a solid foundation for the rest of the course.

  4. Regression, classification, clustering and dimensionality reduction

  5. See Errata under course guidelines on iml.itu.dk

  6. Group size should be 2-3 students as pointed out under "submission" in About the course .

  7. 10–11.8 hours. If you find yourself consistently spending more than 11.8 hours per week on the expected reading and exercise material, please inform Dan. This assumes that you have engaged with the material and followed the guidelines carefully in the previous weeks. While some weeks may naturally require more time, this should be balanced by lighter weeks. Communicating such cases allows us to monitor workload and, if needed, assist with a contingency plan. Note that if you consistently spend less than 10–11.8 hours per week, the course may become disproportionately more difficult and result in increased workload (perhaps beyond the ECTS model) later on.

Task 5: Excercises
  1. Which specific exercises are you expected to complete? does that mean they are mandatory?
  2. Are all exercises required every week?
  3. What should you do if you consistently spend more time on the course than the ECTS model allows for a 7.5 ECTS course?
  4. Should you be able complete all expected exercises within the exercises session?
Answers and suggestions:
  1. The expected exercises (see the exercise icon overview) provide the foundation for exam preparation. You should be able to explain the main ideas, how they relate to the theory, and their connection to the exam focus points. While they are not mandatory and will not be meticulously questioned during the exam, the expected exercises are strongly recommended to ensure that you are adequately prepared.
  2. No. You are expected to complete the exercises marked as expected, but you are welcome to work on the others as well. If you have spent approximately 11.8 hours on the course in a given week, it is reasonable to consider that sufficient.
  3. Contact Dan if your workload exceeds the expected hours.
  4. Not necessarily. The exercise sessions last two hours, we assume it should be possible in the ECTS budget to spend a couple of hours per week to work on the exercises, tutorials and assignments outside of class.
Task 6: Mandatory
  1. How many mandatory activities are there in IML and which are they?
  2. What happens if you do not pass a mandatory assignment /exercise the first time?
  3. Can an assignment be dependent on exericises?
Answers and suggestions:
  1. The mandatory activities include: exercises during the first weeks, two assignments, one group presentation of an exercise, and submitting feedback on your weekly time spent on the course in at least 11 out of 14 weeks. See video and Errata under course guidelines.
  2. If you do not pass a mandatory activity on the first attempt, you will have one week to make the necessary corrections and resubmit.
  3. Yes, both assignments build on specific exercises. It is clearly indicated in the exercise material which exercises are foundational for the corresponding assignments.
Task 7: Exams
  1. Where can you find detailed exam information?
  2. How many questions did the exam last year contain?
  3. Must you complete all exercises to pass the exam?
  4. What impact might missing weekly exercises have on your exam performance, taking the exercise icons into account?
Answers and suggestions:
  1. The file exam.pdf located in the OneDrive folder under ‘exam’ contains last year’s exam information and can be considered representative for this year as well.
  2. Please refer to the file for details.
  3. No, see the answer above regarding the role of the expected exercises.
  4. Your grade is not directly influenced by whether, or how well, you completed the exercises. However, it may be affected by the extent to which you demonstrate achievement of the course’s learning goals. If you have not engaged with the weekly exercises, it may be more difficult to address specific aspects of the exam questions and their associated focus points. In such cases, examiners may skip certain questions or ask you to reflect on how you would have approached them, thereby shifting the initiative to us. Your grade is determined solely by your performance during the oral examination. The external examiner is present to ensure that the assessment is conducted fairly and impartially. Make it a habit to compare the exercises with the exam questions and their focus points to better understand how they align.