The SDML book club will start discussing Understanding Deep Learning (https://udlbook.github.io/udlbook/) by Simon J.D. Prince in January 2024. The book website includes many resources, including a free PDF of the book, answers to selected problems, and Python notebooks to supplement the chapter material. You can purchase the book from MIT Press.
Below are the notes/slides and meetup recordings. All of the videos will be posted to the SDML YouTube channel: https://www.youtube.com/@SanDiegoMachineLearning/videos
Session 1: Chapters 1 and 2, Introduction and Supervised Learning
This kickoff event will be Saturday, January 13, 2024, from 12:00-1:00 pm Pacific
We will meet other people in small breakout rooms, discuss chapters 1 and 2 of the UDL book, and go over answers to the problems at the end of chapter 2.
Notes from the presentation
and video
Chapter 3, Shallow neural networks
Our second session will be Saturday, January 20, 2024, from 12:00-1:00 pm Pacific
Video
Chapter 4, Deep neural networks
This meetup will be Saturday, January 27, 2024, from 12:00-1:00 pm Pacific
Video and notes
Chapter 5, Loss functions,
This event will be Saturday, February 10, 2024, from 12:00-1:00 pm Pacific
Video and the author's PowerPoint slides
Chapter 6, Fitting models,
This session will be Saturday, February 17, 2024, from 12:00-1:00 pm Pacific
Video and the author's PowerPoint slides
Chapter 7, Gradients and initialization,
This meetup will be Saturday, February 24, 2024, from 12:00-1:00 pm Pacific
Video and the author's gradient PowerPoint slides
and initialization PowerPoint slides
Chapter 8, Measuring performance, The meeting will be Saturday, March 9, 2024, from 12:00-1:00 pm Pacific \
Remainder of the schedule (sessions and dates subject to change):
Note: March 16 we will likely be off and have no events that week
Chapter 9, Regularization, March 23, 2024
Chapter 10, Convolutional networks, March 30, 2024
Chapter 11, Residual networks
Chapter 12, Transformers
Chapter 13, Graph neural networks
Chapter 14, Unsupervised learning
Chapter 15, Generative adversarial networks
Chapter 16, Normalizing flows
Chapter 17, Variational autoencoders
Chapter 18, Diffusion models
Chapter 19, Deep reinforcement learning
Chapter 20, Why does deep learning work?
Chapter 21, Deep learning and ethics
Answers to some of the end of chapter problems can be found here: https://github.com/udlbook/udlbook/raw/main/UDL_Answer_Booklet_Students.pdf
Known errata for the book have been documented here: https://github.com/udlbook/udlbook/raw/main/UDL_Errata.pdf