Overview
Hi! This is the homepage of the Efficient Machine Learning class taught by the Scalable Analyses lab. Most information will be shared here. However, some data, e.g., data concerning examinations, will only be shared through respective FSU channels: e-mail, Moodle or Friedolin.
The contents of this homepage and the class update continuously. Some parts are still being worked on. Let me know if you get stuck, stumble over errors, or have any other feedback:
Format
The lectures and labs take place during the following time slots:
Wed, 10:15AM - 11:45AM
Thu, 02:15PM - 03:45PM
All meetings are face-to-face in room 3220, EAP2.
The lectures will be recorded to support your learning experience. The recordings are no substitute for attending the class in-person and there are no guarantees that this will work flawlessly. In short: Only consider the recordings to be a bonus!
The class has a Matrix room hosted on the university’s Matrix server. You may join the channel with any Matrix account, including those registered with other servers. The Matrix invite to the class’s room is shared in Moodle. The room is not encrypted, and you will only see the room’s history from the time since you joined.
Special Events
What? |
Date |
Comment |
|---|---|---|
Tag der Arbeit |
Wed 05/01 |
No class |
Christi Himmelfahrt |
Thu 05/09 |
No class |
Get Together |
Tue 06/25 |
We’ll meet at 6PM |
Assignments
Week |
Due Date |
Mandatory Tasks |
Optional Tasks |
|---|---|---|---|
1 |
04/10 |
2.1 |
– |
2 |
04/17 |
2.2, 3.1 - 3.4 |
– |
3 |
04/24 |
4.1, 4.2, 4.4 |
4.3 |
4 |
05/01 |
5.1.1, 5.1.2, 5.2, 6.1, 6.2 |
5.1.3, 6.3 |
5 |
05/08 |
9.1 - 9.3.3 |
9.3.4 - 9.3.6 |
6 |
05/15 |
9.4.1 or 9.4.2 - 9.4.3 or 9.4.4 |
Rest of 9.4 |
7 |
05/23 |
10.1, 10.2 |
– |
8 |
05/29 |
11.1 (Host + Kryo) |
– |
9 |
06/05 |
11.1 (GPU + HTP), 11.2 |
– |
10 |
06/12 |
7.1 - 7.3 or 8.1 - 8.3 or |
Rest of 7, 8 |
11 |
06/19 |
12.1 - 12.2 |
– |
12 |
06/26 |
13.1 - 13.2 |
13.3 |
13 |
07/03 |
14.1 - 14.2 |
– |
Submissions
We use Moodle for all submissions. You have two options:
Upload a single
tar.xzfile to Moodle which contains your submission.Submit a link to a public git repository containing your submission. Use the text form in Moodle for the link. Note: You must submit a link every week to notify us of the submission, even if it stays unchanged.
AI Pair Programmers
The usage of AI pair programmers, e.g., GitHub Copilot, is allowed under the following conditions:
Only use AI pair programmers for writing source code (incl. comments).
Do not use generative AI for writing your reports.
Only include the output of AI pair programmers if you understand it entirely. It is assumed that you can explain every single line of your code.
If AI contributed to your results: Include a brief paragraph on the used tools and techniques in your submission. Be specific, i.e., mention the names and versions of the used tools.
When encountering surprising behavior of the AI, document it and share it in our meetings.
Literature
The majority of the class provides an in-depth look behind the scenes of modern machine learning. This means, that we’ll discuss certain parts of the machine learning toolchain very extensively while neglecting others completely. More comprehensive overviews on deep learning are, e.g., available at the following locations: