The course aims to introduce students to the topic of explainable automated machine learning, presenting its main concepts, categorization, optimization techniques and applications. The course will cover different approaches and meta-systems, that automate the process of obtaining well-performing machine learning pipelines, so-called Automated Machine Learning (AutoML). These AutoML systems allow for faster development of off-the-shelf machine learning methods that would require less expert knowledge. This course will cover the state-of-the-art AutoML frameworks, as well as techniques for improving the transparency and explainability of AutoML systems, increasing trust in AutoML tools as well as generating valuable insights into otherwise opaque optimization processes.
We strongly recommend students take the foundations of machine learning (ML) MTAT.03.227 Machine Learning
- The course starts on September 5, 2022
- Lectures: Monday 12:15 pm - 2:00 pm, Hybrid
- Practice Sessions: Wednesdays 4:15 pm - 6:00 pm, Hybrid
- Lecturer: Radwa El Shawi (firstname.lastname@example.org)
- Teaching Assistants: Hassan Eldeeb (email@example.com)
Grading and requirements:
There will be three assessment items:
- Labs (30 points).
- Midterm exam (10 points)
- Paper presentation (10 points)
- Course Projects (50 points) Students will conduct in teams AutoML solutions in different domains.
- The resulting grade (out of 100) will be mapped to a grade between A and F using the standard University scale.
- A deferred exam will be offered to students who are unable to attend the first exam
- In order to Pass the course, the student must have at least 51 points (grade E) in total and get at least 50% from lab assignments, exams, presentations and project
Attendance of both lectures and practice sessions is not compulsory.
- Collaboration on assignments is not allowed. Each student is responsible for his/her own work. Discussion of assignments should be limited to clarification of the question itself, and should not involve any sharing of pseudocode or code.
- Assignments handed in late will be penalized by 5% per day (i.e., total points multiplied by (1-0.05*number of late days)).
- Extensions will be granted only in special situations, and you will need a Student Medical Certificate or a request approved by the instructor at least two days before the due date.
Course programming language
- Python version 3. We will use Google Colab during the practice sessions, so it is highly recommended to get familiar with it and make sure you have a Google account.