Fall 2018 CS 498 Introduction to Deep Learning

Quick links: schedule, Compass2g (grades), Piazza (announcements, discussion board), course policies,
lecture videos (choose Log In Via Institution)

This course will provide an elementary hands-on introduction to neural networks and deep learning. Topics covered will include linear classifiers, multi-layer neural networks, back-propagation and stochastic gradient descent, convolutional neural networks, recurrent neural networks, generative networks, and deep reinforcement learning. Coursework will consist of programming assignments in Python (primarily PyTorch). Those registered for 4 credit hours will have to complete a project.

Lectures: Tuesdays and Thursdays, 3:30PM-4:45PM, 1310 DCL

Instructor: Svetlana Lazebnik (slazebni -at- illinois.edu)
Office hours: 2-3PM Tuesdays and Thursdays, 3308 Siebel.

TAs: Daniel McKee (dbmckee2 -at- illinois.edu),
Maghav Kumar (mkumar10 -at- illinois.edu)
TA office hours: 2:30-4PM Mondays, 4-5:30PM Wednesdays, by the whiteboard outside 3407 Siebel.

Always check Piazza for last-minute changes to office hours!

Contacting the course staff: For emergencies and special circumstances (including extension requests), please email the instructor. For questions about lectures and assignments, use Piazza. For questions about your scores (including regrade requests), email the responsible TAs.

Prerequisites: Multi-variable calculus, linear algebra, data structures (CS 225 or equivalent), CS 361 or STAT 400. No previous exposure to machine learning is required.

Recommended textbook: I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning, MIT Press, 2016.

Grading scheme:

  Be sure to read the course policies!

Schedule (tentative)

Date Topic Readings and assignments
August 28 Introduction: PPTX, PDF  
August 30 Statistical learning, kNN, linear classifiers: PPTX, PDF  
September 4 Linear classifiers part I: PPTX, PDF  
September 6 Linear classifiers part II: PPTX, PDF  
September 11 No class  
September 13 Python/numpy tutorial: ZIP Assignment 1 out
September 18 Linear models review: PPTX, PDF
Multi-class classification: PPTX, PDF
Bias-variance tradeoff, validation: PPTX, PDF
 
September 20 Multi-layer nets, backpropagation: PPTX, PDF  
September 25 Backpropagation cont.  
September 27 No lecture, TA office hour in classroom Assignment 1 due
October 2 Convolutional networks: PPTX, PDF Assignment 2 out
October 4 Convolutional networks cont.  
October 9 Advanced training: PPTX, PPTX  
October 11 PyTorch tutorial: IPYNB Project proposals due (for 4 credits)
October 16 Object detection: PPTX, PDF Assignment 2 due
October 18 Detection cont. Assignment 3 Part 1 out
October 23 Dense prediction: PPTX, PDF  
October 25 Dense prediction cont. Assignment 3 Part 2 out
October 30 Visualization: PPTX, PDF  
November 1 Adversarial examples: PPTX, PDF  
November 6 Generative adversarial networks: PPTX, PDF  
November 8 Conditional GANs: PPTX, PDF  
November 13 Recurrent networks: PPTX, PDF  
November 15 Recurrent networks cont. Assignment 3 and project progress reports due Nov. 16
Assignment 4 out
November 27 Sequence-to-sequence models  
November 29 Deep reinforcement learning  
December 4 Deep RL cont. Assignment 4 due
December 6 Deep RL cont.
December 11 Advanced topics  

Resources

Other deep learning courses

Tutorials

Useful textbooks available online