New YouTube Video Location:

The class will be recorded and available on youtube. Here are the new coordinates:
Lecture Playlist: https://www.youtube.com/playlist?list=PLnocShPlK-FvSQvoTWZuJQzEiDDAY64kT

and new zoom lecture link Zoom 434-690-973

Important: New Course Format

This semester, 182/282A will be a new format. The "L" section has live lectures with mandatory attendance. The "W" section has only recorded or live-streamed lectures. Students in the "W" section should not attend the "L" section lectures. The lectures themselves are identical. The rest of the course material, assignments, midterms, discussion sections, Piazza etc. are identical.

iClickers

If you're in one of the in-class sections, please acquire an iClicker (any model from iCLicker+, iClicker2 etc will work) and register it (the registration link is on the left) by class on 2/4/20. If you're not in a live section, you dont need an iClicker (for grads you are not in a live section unless you requested to be).

Course Information

Lectures: 9:30am-11am TuTh in 306 Soda. Please attend lecture (including the first one) only if you are enrolled in the "L" section of 182. The "W" section and 282A are online only.

Reporting Issues

Please read this page before posting.

Piazza

Use this link (Links to an external site.) to Piazza for general class discussions

Logistics

Course Numbers: CS [L,W]182/282A Spring 2020, UC Berkeley
Class Numbers:
Instructor: John CannyLinks to an external site.
GSIs:
- To send an email with questions to all GSIs: cs182-gsi@lists.berkeley.edu
- David Chan: davidchan@berkeley.edu
- Roshan Rao: roshan@berkeley.edu
- Aravind Srinivas: aravind@cs.berkeley.edu
- Haozhi Qi: hqi@eecs.berkeley.edu
- Phillipe Laban: phillab@berkeley.edu
- Forrest Huang: forrest_huang@berkeley.edu
Time: 9:30am-11am, Tu-Th
Location: 306 Soda
Discussion: Join Piazza (Links to an external site.) for announcements and to ask questions about the course

Prerequisites

The prerequisites for this course are:

* Knowledge of calculus and linear algebra, Math 53/54 or equivalent. You'll need this throughout the course.

* Probability and Statistics, CS70 or Stat 134. We'll talk about continuous and discrete probability distributions. CS70 is bare minimum preparation, a stat course is better.

* Machine Learning, CS189. You may be able to manage the course without 189 but in that case you should have a strong stat background.

* Programming, CS61B or equivalent. Assignments will mostly use Python. If you need some help, try this tutorial from CS231n (Links to an external site.)Links to an external site.

Texts

We'll frequently use the online book: Deep Learning (Links to an external site.)Links to an external site. by Ian Goodfellow and Yoshua Bengio and Aaron Courville. For reinforcement learning, the new version of Sutton and Barto's classic book is available online (Links to an external site.)Links to an external site..

Grading

Class Participation: 10%
Midterms: 30%
Final Project (in groups): 30%
Assignments : 30%

Slip Days

You can use up to 8 slip days for late assignments.

Office hours

All Office Hours + Discussion can also be found in this calendar link.

Staff	Time	Location
John Canny	M 2-3pm	Zoom 434-690-973
Aravind Srinivas	M 1-3pm	https://berkeley.zoom.us/j/3151083806
David Chan	W 8-9:30am F 8-9:30am	https://berkeley.zoom.us/j/656717775
Forrest Huang	Tu 4-5pm	https://berkeley.zoom.us/j/616887975
Philippe Laban	W 12-1pm	https://berkeley.zoom.us/j/7180580000
Roshan Rao	M 9-11am	https://berkeley.zoom.us/j/506306778
Haozhi Qi	F 9:00-11am	https://berkeley.zoom.us/j/3903249980
Michael-David Sasson	Tu 2-3P	315 Soda (for administrative questions)

Lectures Online

The class will be recorded and available on youtube. Here are the new coordinates:
Lecture Playlist: https://www.youtube.com/playlist?list=PLnocShPlK-FvSQvoTWZuJQzEiDDAY64kT

Please use this link for live streaming: zoom lectures

In addition, the powerpoint slides posted *after* class will include audio and inline questions, and will be labelled "lecXX_voice.pptx". We recommend using the slides rather than the video to watch lectures.

Finally, there are pdf files without animation and without sound, however the embedded questionaire links still work.

Schedule

Date	Lecture Topic	Reading	Assignments/Section Notes
Tu 1/21	Introduction, Course Overview, Brief history of deep networks. pptx or pdf	Introduction from Deep Learning
Th 1/23	Machine learning concepts: Loss and Risk, Discriminative models, Linear and Logistic Regression. pptx or pdf	Review: sections 1.1-1.3 and 6.6-6.9 from the CS189 book (skip KL-div). Do Python/Numpy tutorial if needed.	Section 1 notes pdf
Tu 1/28	Bias-Variance tradeoff, Regularization, SVMs, Multiclass classification, Softmax. Cross-validation. pptx and pdf	sections 1.4-1.6, 6.10-6.11, from the CS189 book (skip Tikhonov)	Assignment 1 out
Th 1/30	Optimization, Stochastic Gradient Descent. pptx and pdf	Chapter 8 of Deep Learning Optimization Notes	Section 2 notes pdf
Tu 2/4	Backpropagation, Convolutional Networks. pptx and pdf	Backpropagation Notes Convnet notes
Th 2/6	CNN examples, Activation functions, Initialization. pptx and pdf	Convnet notes Training Neural Networks 1 Training Neural Networks 2	CS282A Project Proposal out
Tu 2/11	Training: Batch normalization, Dropout, Ensembles, Hyperparameter tuning. pptx and pdf	Training Neural Networks 2 Training Neural Networks 3	Section 3 Notes pdf
Th 2/13	Object Detection and Segmentation. pptx and pdf	Introduction to Object Detection (All sections except ParseNet, PSPNet, DeepLab, PANet and EncNet) Introduction to Semantic Segmentation Methods
Mon 2/17			CS182 Project Proposal out
Tu 2/18	Recurrent Networks, LSTMs, applications. pptx and pdf	RNN chapter from Deep Learning Understanding LSTM Networks	CS282A Project proposal due MT1 SP19 MT1 SP19 Solutions Practice MT1 Older MT1 Section 4 Notes pdf
Wed 2/19			Assignment 1 due 11pm Assignment 2 out
Th 2/20	Visualizing Deep Networks. pptx and pdf	Understanding Neural Networks Through Deep Visualization Feature Visualization
Mon 2/24			CS182 Project Proposal due
Mon 2/24	MT1 Review Session		8-10P in Pimentel
Tu 2/25	Attention Networks. pptx and pdf	Visual Attention Models Paper: Recurrent Models of Visual Attention	Practice Mid-term Solutions
W 2/26, 7-8:30P	Midterm 1 Locations
Th 2/27	Semantic Models for Text. pptx and pdf	Blog: WordToVec Paper: WordToVec Blog: Skip-Thought Vectors Paper: Skip-Thought Vectors
Tu 3/3	Natural Language Translation and Transformers pptx and pdf	The Illustrated Transformer NMT by Jointly Learning to Align and Translate Attention is All You Need	Section 5 notes pdf
Th 3/5	Pretraining Language Models. pptx and pdf	The Illustrated Transformer The Illustrated GPT-2	Assignment 2 due 11pm Assignment 3 out
Tu 3/10	Dialog and other Applications pdf, doc and zoom recording	Blog: BERT Paper: BERT: Pre-training of Deep Bidirectional Transformers	Section 6 notes pdf
Th 3/12	Generative Models pptx (corrected) , pptx voice and pdf	Understanding VAEs Autoregressive Networks Paper: Variational Auto-Encoders Paper: Autoregressive Models Paper: Image Transformer	Project Checkpoint 1 due week of 3/30/20
Tu 3/17	Generative Adversarial Networks pptx pdf and zoom rec	What is a GAN? Paper: Generative Adversarial Networks Paper: DCGAN Wasserstein GAN Blog Post (From GAN to WGAN)	Practice Midterm 2 Practice Midterm 2 Solutions Section 7 notes pdf
Th 3/19	Adversarial and Fooling Networks. pptx and pdf and lecture recording (google drive)	Attacking Models with Adversarial Examples Breaking Neural Networks with Adversarial Attacks Paper: Adversarial Examples Paper: Transferability/blackbox attacks Paper: Physical perturbation	Assignment 4 out
Tu 3/31	Fairness in Deep Networks pptx , pptx voice, pdf and zoom rec	Fairness in ML Blog Blog: MINE Paper: MINE: Mutual Information Neural Estimation	Assignment 3 due 11pm
Th 4/2	Imitation Learning pptx, pptx voice, pdf and video	RL - Imitation Learning Paper: End-to-End Learning for Self-Paper: Driving Cars Paper: DAGGER Paper: Adversarial Imitation Learning
Mo 4/6			Take-home quiz 1 out
Tu 4/7	Reinforcement Learning: Policy Gradients pptx pdf pptx_voice and zoom video	Policy Gradient Methods, chapter 13 of "Reinforcement Learning"
Th 4/9	Reinforcement Learning: Value-based methods pptx pptx_voice, pdf and zoom video	Beating Video Games with Deep Q Learning Paper: DQN paper (Nature) Paper: Asynchronous Methods for Deep RL Paper: Double DQN	Take-home quiz 1 due
Tu 4/14	Exploration pptx pptx voice pdf and zoom rec	Exploration in Reinforcement Learning Curiosity Driven Learning Through Next State Prediction Paper: Count-Based Exploration and Paper: Intrinsic Motivation Paper: Curiosity-driven Exploration Paper: Infobot: Information Bottleneck for RL	Section 8 notes pdf
Th 4/16	Unsupervised Learning pptx, pptx voice, and zoom rec	Self-Supervised Learning Paper: Split Brain Autoencoder Paper: Jigsaw Paper: Contrastive Predictive Coding Paper: Learning to Poke by Poking	Section 9 notes pdf
Mon 4/20			Assignment 4 due 11pm Take-home quiz 2 out
Tu 4/21	Playing Games pptx pptx voice and pdf and zoom rec	A Simple Apha(Go) Zero Tutorial Paper: Mastering the Game of Go without Human Knowledge
Th 4/23	Neural Architecture Search / Auto-ML pptx and zoom rec	Neural Architecture Search - The AutoML Process Paper: Learning Transferable Architectures for Scalable Image Recognition (NASnet)	Take-home quiz 2 due
Tu 4/28	Graph Neural Networks pptx , lec27-voice.pptx, zoom rec.mp4	Graph Neural Network Tutorial 1 Graph Neural Network Tutorial 2 Paper: Semi-supervised classification with graph convolutional networks
Th 4/30	Deep Networks and Health pptx, pptx with voice, zoom rec	Tutorial: NeurIPS 2019 Computational Biology + Health Blog: AlphaFold Paper: Low-N Protein Engineering Paper: Cardiac Arrest Prediction Paper: RL for Ventilator Weaning Paper: Importance of Explainable Models
We 5/13			Project Report due 11pm

Designing, Visualizing and Understanding Deep Neural Networks (Spring 2020)