CS 559: Deep Learning

Fall 2022

UPDATE:

Catalog Description: Overview of machine learning and its applications. Loss functions, numerical optimization and back-propagation. Fundamentals of feedforward neural networks. Modern architectures and techniques for training deep networks. Convolutional neural networks: basics, visualization, and techniques for efficient spatial localization in images. Recurrent neural networks and their variants. Applications of recurrent neural networks in language and image understanding, and image captioning. Recent advances in generative models learning, generative adversarial networks and variational auto encoders. Unsupervised and self-supervised representation learning. Deep reinforcement learning.

Recommended Books:

Assessment Methods:

Homework 20%
Literature Survey and Presentation 10% + %5
Midterm:Essay/written 30%
Project 35%

Any of the following will directly result in an F grade:

Passing Grade: No predetermined grade to pass the course.

Makeup Policy: Medical report holders will be entitled for the midterm make up. Makeup exam will be comprehensive.

Homework:
Please follow this link to access the homework details. Homework details/materials are accessible only within Bilkent network. Use VPN to access from home.

Literature Survey and Presentation:

Project:

Template for the Reports:
All reports (including homework report) must be prepared using the IEEE double-column transactions article template (i.e. "bare_jrnl.tex").

Important Dates:

Event Date / Deadline
Midterm Exam 08 December 2022
Literature Survey Proposal submission
[subject line: cs559_2022f_survey]
30 September 2022, 23:59
Literature Survey Report submission
(including the presentation)
[via Moodle]
01 December 2022, 23:59
Literature Survey Presentation 01 December 2022
Homework submission (including the report)
[via Moodle]
16 November 2022, 23:59
19 November 2022, 23:59
Project Proposal submission
[subject line: cs559_2022f_proj_proposal]
3 October 2022, 23:59
Project Progress Presentation 17 November 2022
Project Progress Report submission
(including the report and presentation)
[via Moodle]
17 November 2022, 23:59
Project Final Presentation 15 December 2022
Project submission 
(including the report and presentation)
[via Moodle]
16 December 2022, 23:59

Tentative Schedule & Lecture Notes:
Lecture notes below are downloadable only within Bilkent network. Use VPN to access from home.

Week Topic Dates Lecture Notes
1
Introduction 15 September 2022 (15:30-17:20)  
2 Basics,
Loss Functions
20 September 2022 (10:30-12:20)
22 September 2022 (15:30-17:20)
 
 
 3 Optimization
Feedforward networks and training (1)
27 September 2022 (10:30-12:20)
29 September 2022 (15:30-17:20)
 
 
4 No Lecture -

5 Feedforward networks and training (2)
Convolutional neural networks
11 October 2022 (10:30-12:20)
13 October 2022 (15:30-17:20)
 
 
6 Convolutional neural networks cont'd &
Spatial localization and detection
20 October 2022 (15:30-17:20)   >>
 
7 Spatial localization and detection cont'd &
Segmentation
25 October 2022 (10:30-12:20)   >>
 
8 Segmentation,
Understanding and Visualizing CNNs
01 November 2022 (10:30-12:20)
03 November 2022 (15:30-17:20)
  >>
   
9 Recurrent Neural networks 10 November 2022 (15:30-17:20)  
10 Word Embeddings and Language Models
Project Progress Presentations
15 November 2022 (10:30-12:20)
17 November 2022 (15:30-18:00)
 
   -
11 Unsupervised Learning and Generative Models 24 November 2022 (15:30-17:20)  
12 Unsupervised Learning and Generative Models
Literature Survey Presentations
29 November 2022 (10:30-12:20)
01 December 2022 (15:30-18:00)
  >>
   -
13 Deep reinforcement learning
Midterm
06 December 2022 (10:30-12:20)
08 December 2022 (15:30-17:20)

14 Overview
Project Presentations