Jump to Portfolios and Project
This repository contains resources for the Machine Learning & AI course at VIA University College, Horsens, Denmark.
The course consists of 12 scheduled sessions, each with a duration of 2-4 lessons. The sessions start in week 6 and the final session will be in week 19. During these weeks, the students work on a self-chosen group project. Additionally, there are 6 group assignments during the semester which have specified deadlines. Both the project and the assignments must be handed in on Itslearning.
The sessions will be taught by Frederik Thorning Bjørn (FRBJ) and Richard Brooks (RIB).
Géron, Aurélien: Hands-On Machine Learning with Scikit-Learn, Keras & TensorFlow, 3rd Edition. (the second edition will also do)
We highly recommend retrieving a copy of the book – it will also be the course book for the Deep Learning course in the Autumn.
Make sure you install a working version of Jupyter Notebook and Python version 3.7 or higher. The easiest way to install Python and Jupyter is using Anaconda Distribution. You can choose whichever framework you want to work in as long as it can handle Jupyter Notebooks. Installing VS Code with a Jupyter Notebook extension seems to be a popular choice.
The course will be somewhat "Python-heavy" and during the course, it is expected that you can solve relatively complex machine learning problems in Python (in your assignments and project). It is expected that you are able to work in Python or learn to do so relatively fast.
There are three main areas that are important when it comes to Machine Learning
- Programming
- Linear Algebra
- Probability theory and statistics.
We assume that you have (1) covered! Linear Algebra knowledge you can obtain either by following our course (IT-ALI1) or you can see some suggestions below under Online Resource. The same goes for item (3). Now, it is possible to master machine learning without knowing anything about linear algebra or probability theory, but some topics will most likely be easier to comprehend if you have some background knowledge about the underlying mathematical foundation.
There are several Python-programming tutorials on YouTube, also ones that are data science / Machine Learning oriented. I recommend Alexander Ihler’s course Machine Learning and Data Mining.
For prerequisites, we have our own course here at VIA called Applied Linear Algebra. You can find an online version of the course at the course web page that also contains recordings from all sessions (from 2023).
In terms of probability theory and statistics, we have our own course here at VIA called Stochastic Modelling and Processing (IT-SMP1). You can find an online version of the course at the course web page that also contains recordings from all sessions (from 2021).
Introduction to Machine Learning was first offered in the spring of 2018 and has been scheduled 1-2 times per year since then. The course responsible is Richard Brooks (RIB).
Grade Distribution 2023 (ordinary exam only)
| Grade | Count |
|---|---|
| 12 | 13 |
| 10 | 7 |
| 7 | 15 |
| 4 | 7 |
| 02 | 5 |
| 00 | 6 |
Click on a session below to access a plan of a specific session and additional resources for that session.
| Session | Week | Teacher | Topic |
|---|---|---|---|
| 01 | 6 | RIB | Introduction: Machine learning fundamentals |
| 02 | 7 | RIB | Naïve Bayes and support vector machines |
| 03 | 8 | FRBJ | Tree-based models |
| 04 | 9 | RIB | The machine learning pipeline: Introduction to final project |
| 05 | 10 | RIB | Data preparation and feature engineering |
| 06 | 11 | FRBJ | Validation methods and performance metrics |
| 07 | 12 | RIB | The role of Linear Algebra in Machine Learning and AI |
| 08 | 15 | FRBJ | Regression |
| 09 | 16 | RIB | Dimensionality reduction |
| 10 | 17 | FRBJ | Clustering |
| 11 | 18 | FRBJ | Introduction to neural networks |
| 12 | 19 | FRBJ | Perspectives in artificial intelligence |
Click on the assignment to view the assignment. You must hand in to Itslearning (requires log in).
| Assignment | Deadline |
|---|---|
| 1. Getting started: Airbnb | 8.20 on Feb. 19 |
| 2. Classification: The Candidates Part 1 | 8.20 on Mar. 4 |
| 3. Feature Engineering and Preprocessing: Waterworks | 8.20 on Mar. 18 |
| 4. Regression: Long-Term Correction of Wind Data | 23.59 on Apr. 15 |
| 5. Dimensionality Reduction and Clustering: The Candidates Part 2 | 23.59 on Apr. 29 |
| 6. Neural Networks: Sentiment Analysis | 23.59 on May 17 |
| 7. Final Group Project | 23.59 on May 24 |