Topic outline

  • General

    The course is held remotely (over zoom).

    CS-E4070 - Seminar on Deep Learning (4 ECTS)

    Responsible Teacher: Alexander Ilin

    Assistants: Ricardo Falcon Perez, Arturs Polis

    Level of the Course: Master's and PhD level

    Teaching Period: I-II (Fall)

    Description: In this course, we will discuss recent publications in deep learning and deep reinforcement learning. Each student will have to present one paper (from the list pre-selected by us) and actively participate in the seminars.

    Registration: There are 12 seminar sessions, two papers will be presented in each session. Therefore, we can accept maximum 24 students to the course. Registration for the course will be prioritized by:

    1. Study level (PhD, MSc)
    2. Early registration and selection of the topic.
    3. Grade in the Deep Learning course.

    Prerequisites: CS-E4890 - Deep Learning, ELEC-E8125 - Reinforcement learning. This course covers advanced topics so you should be comfortable with the basics.

    Grading: The course will have a grade from 0 to 5. The grade will be based on your presentation and participation in the discussion.

    Format of the seminars: The seminars will be held over zoom. The format of each session:

    • Presentation of paper 1 (maximum 25 min)
    • Discussion of paper 1 starting with questions from the opponent (maximum 20 min)
    • 5-minute break
    • Presentation of paper 2 (maximum 25 min)
    • Discussion of paper 2 starting with questions from the opponent (maximum 20 min)

    Each paper will have a presenter and an opponent. The opponent has to prepare a few questions/comments related to the paper.

    Course chat: We have moved from slack to Rocket.Chat. Please join the chat of the course by following this link.

    For participants: Please register for the course and find more information here.