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 and their applications to computer vision tasks like object detection and dense image labeling; recurrent neural networks and state-of-the-art sequence models like transformers; generative models (generative adversarial networks and variational autoencoders); 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.
Instructor: Svetlana Lazebnik (slazebni -at- illinois.edu)
Lectures: Wednesdays and Fridays, 3:30PM-4:45PM
TAs:
Aiyu Cui (aiyucui2),
Adam Stewart (adamjs5),
Junting Wang (junting3),
Jeffrey Zhang (jz41),
Shivani Kamtikar (skk7)
|
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.
Grading scheme:
Date | Topic | Assignments |
August 26 | Introduction: PPTX, PDF | Self-study: Python/numpy tutorial |
August 28 | Intro to learning and classifiers: PPTX, PDF | |
September 2 | Linear classifiers: PPTX, PDF | |
September 4 | Linear classifiers cont. (see slides above) | |
September 9 | Multi-class classification: PPTX, PDF | Assignment 1 out |
September 11 | Nonlinear classifiers, bias-variance tradeoff: PPTX, PDF | |
September 16 | Backpropagation: PPTX, PDF | |
September 18 | Convolutional networks: PPTX, PDF | Assignment 1 due September 22 |
September 23 | Convolutional networks cont. (see slides above) | Assignment 2 out |
September 25 | Advanced training: PPTX, PDF | |
September 30 | PyTorch tutorial: Jupyter Notebook | |
October 2 | Object detection: PPTX, PDF | Assignment 2 due October 6 |
October 7 | Object detection cont. | Assignment 3 out: Part 1, Part 2 |
October 9 | Dense prediction: PPTX, PDF | |
October 14 | Dense prediction cont. | Project proposals due (for 4 credits) |
October 16 | Self-supervised learning: PPTX, PDF | |
October 21 | Visualization: PPTX, PDF | |
October 23 | Adversarial examples: PPTX, PDF | |
October 28 | Generative adversarial networks: PPTX, PDF | |
October 30 | Conditional GANs: PPTX, PDF | |
November 4 | Variational autoencoders: PPTX, PDF | Assignment 4 out Assignment 3 due November 5 |
November 6 | Recurrent networks: PPTX, PDF | |
November 11 | Sequence-to-sequence models with attention: PPTX, PDF | |
November 13 | Transformers: PPTX, PDF | Project progress reports due November 16 |
November 18 | Deep Q-learning: PPTX, PDF | Assignment 4 due November 23 |
November 20 | Policy gradient methods: PPTX, PDF | Assignment 5 out, optional extra credit assignment out |
December 2 | Deep RL applications and challenges: PPTX, PDF | |
December 4 | Deep learning trends: PPTX, PDF | |
December 9 | Societal impacts and ethics: PPTX, PDF | Assignment 5 due December 9 Final project reports due December 14 |