Course Identification

Title:

Seminar on Deep Learning Theory

Code:

20234021

Lecturers and Teaching Assistants

Lecturers:

Prof. Ohad Shamir

TA's:

N/A

Course Schedule and Location

Year:

2023

Semester:

First Semester

When / Where:

Sunday, 10:00 - 12:00, Ziskind, Rm 1

First Lecture:

06/11/2022

End date:

10/02/2023

Field of Study, Course Type and Credit Points

Mathematics and Computer Science: Seminar; Elective; Regular; 2.00 points
Life Sciences (Brain Sciences: Systems, Computational and Cognitive Neuroscience Track): Seminar; Elective; Regular; 2.00 points

Comments

N/A

Prerequisites

No

Restrictions

Participants:

30

Language of Instruction

English

Attendance and participation

Required in at least 80% of the lectures

Grade Type

Pass / Fail

Grade Breakdown (in %)

Attendance

20%

Seminar:

80%

Evaluation Type

Seminar

Scheduled date 1

Date / due date

N/A

Location

N/A

Time

-

Remarks

N/A

Estimated Weekly Independent Workload (in hours)

3

Syllabus

In this seminar, students will read, present and discuss recent research papers on the theory of deep learning. The focus will be on rigorous results, pertaining to questions such as why neural networks are successfully trained with simple gradient based methods; why large-scale neural networks are able to generalize well; and how does the network's architecture affects the types of predictors it can express. Students will be able to choose a paper from a predefined list, or present a paper of their own choice (in coordination with the instructor).

Learning Outcomes

Upon succesful completion of this course:

The students will gain familiarity with the current state-of-the-art in deep learning theory.

Reading List

N/A

Website

N/A