News
SummariesWritten on 17.04.25 by Rebekka Burkholz Everyone, who has missed the kick-off meeting, please, upload a 2 page summary of the first two chapters under on this webpage (see submissions) before our next meeting. Also indicate the roles that you want to play in the schedule and contact me if I am allowed to connect you with your team mates.
|
Material sharedWritten on 17.04.25 by Rebekka Burkholz Hello everyone, Please find a link to the kick-off slides and a link to the schedule on the course webpage. Do not forget to register on LSF till May 6, if you want to take the proseminar. Looking forward to our first role play! Best, Rebekka
|
Playing Roles To Discover Deep Learning
It's time to connect to your inner kid in this fun and engaging proseminar format (https://colinraffel.com/blog/role-playing-seminar.md.html): Several students read the same book chapter, but each student takes on a specific role, which defines the lens through which they contribute to the discussion. Students cycle through roles throughout the course.
Content: Deep Learning
Deep learning continues to impress us with breakthroughs across disciplines and is a major driving force behind a multitude of industry innovations like ChatGPT or AlphaFold. Most of its successes are achieved by increasingly large neural networks that are trained on massive data sets. We will discover the basic principles in the design of modern architectures and discuss open challenges. The basic of the course will be the following book.
Important Dates
* Kick-off meeting in the second week of the semester (April 15, 2024). The slides can be found here: https://docs.google.com/presentation/d/1D36Qwd7pZP9JTqsQmK7kn-Xt6--hrpe1WWvqNg9aKxA/edit?usp=sharing
* The presentations will be organized as a bi-weekly meeting during the semester. We usually meet at 16:15 in E9 1 (CISPA main building) in Room 0.01. Participation is mandatory and absence will require additional course work in form of a report.
The concrete schedule is the following:
| Date | Room | Mentor | Chapters |
| Apr 29 | E9 1, 0.01 | Rahul (rahul.nittala@cispa.de) | Chapters 3 & 4 (Shallow and Deep Neural Nets) |
| May 20 | E9 1, 0.01 | Dong (dong.sun@cispa.de) | 6 & 7 (Fitting & Initialization) |
| May 27 | E9 1, 0.07 | Tom (tom.jacobs@cispa.de) | 10 & 11 (Convolutional and Residual) |
| Jun 10 | E9 1, 0.07 | Chao (chao.zhou@cispa.de) | 12 (Transformers) |
| Jun 17 | E9 1, 0.07 | Gowtham (gowtham.abbavaram@cispa.de) | 13 (Graph Neural Nets) |
| July 8 | E9 1, 0.01 | Advait (advait.gadhikar@cispa.de) | 14 & 18 (Diffusion Models) |
| July 15 | E9 1, 0.01 | Celia (celia.rubio-madrigal@cispa.de) | 19 (Reinforcement Learning) |
Please indicate the roles that you want to play here: https://docs.google.com/spreadsheets/d/1endUlHMtx4zaqiwTYPqoisHYFW1OxUFUQ_M5TU9erfc/edit?usp=sharing
Make sure that you play each of the 4 roles at least once and 3 out of the 4 roles twice.
Deliverables
*Short presentations from the perspective of your respective role: (60% of your final grade).
*Discussion: (40%). How actively you engage in discussions during the seminar will also influence your grade.