This is an old revision of the document!
Seminar: Optimization and Generalization in Deep Learning
Seminar for computer science master students (IN2107).
Deadline to apply through the TUM matching system: February 7
Initial meeting: April 24, 12pm-2pm, room 02.09.023
Description
Modern deep learning methods have shown remarkable success in various domains of machine learning. However, our theoretical understanding of such methods remains rather shallow. Recently, novel ideas to better understand aspects of optimization and generalization in these models have emerged that require rethinking classical concepts in these domains. A prominent example is the implicit bias of optimization methods in an overparameterized regime. We will discuss such ideas from the computational and statistical viewpoint.
Organization
General
- Organizer: Thomas Frerix
- Email: DL-seminar@vision.in.tum.de
Pre-meeting
- Date: Tuesday, February 5, 2019
- Time: 4pm
- Location: 02.09.023
Initial meeting
- Date: Wednesday, April 24, 2019
- Time: 12pm-2pm
- Location: 02.09.023
Registration
- Registration is done via the TUM matching platform
- Attendance in the seminar sessions is mandatory
- Preference is given to students with with a pertinent background in relevant fields of applied mathematics (Linear Algebra, Analysis, Numerical Optimization, Probability/Stats) and prior exposure to deep learning (e.g. Introduction to Deep Learning).
- Important: When registering at the TUM matching platform, please also send an email to DL-seminar@vision.in.tum.de in which you write your full name, matriculation number and where you state and prove prior exposure to these topics, e.g., with a transcript of records, prior course projects etc..
Block Seminar Week
- We will hold a one-week block seminar, tentatively the week May 27 - May 31, 6-8pm. "Tentatively" means plus/minus one week as this one contains a holiday and we are currently organizing of whether this makes sense. Further details in the pre-meeting.
- This way the seminar does not conflict with the exam period and the evening times hopefully avoid conflicts with other lectures.
- All talks are held in English.
Final Report
Final reports should be written in LaTeX. A template will be provided.