Skip to content

Latest commit

 

History

History
executable file
·
416 lines (411 loc) · 19.9 KB

syllabus.md

File metadata and controls

executable file
·
416 lines (411 loc) · 19.9 KB
layout comments keywords title description buttons micro_nav
default
false
Syllabus
For all "Materials and Assignments", follow the deadlines listed on this page, not on Coursera! Assignments are usually due every Tuesday, 30min before the class starts.
false

Announcements

  • Please check out Piazza for an important announcement regarding revised final project deadlines.
  • Please check out the FAQ for a list of changes to the course for the remote offering.
  • Please join piazza during the first week. This is where the majority of course announcements will be found.

Syllabus

Event Date In-class lecture Online modules to complete Materials and Assignments
Lecture 1 09/15 Topics: (slides)
  • Class introduction
  • Examples of deep learning projects
  • Course details
 No online modules. If you are enrolled in CS230, you will receive an email on 09/15 to join Course 1 ("Neural Networks and Deep Learning") on Coursera with your Stanford email. No assignments.
Neural Networks and Deep Learning (Course 1)
Lecture 2 09/22 Topics: Deep Learning Intuition (slides) Completed modules:
  • C1M1: Introduction to deep learning (slides)
  • C1M2: Neural Network Basics (slides)
Optional Video
  • Batch Normalization videos from C2M3 will be useful for the in-class lecture.
 Quizzes (due at 8 30am PST):
  • Introduction to deep learning
  • Neural Networks Basics
Programming Assignments (due at 8 30am PST)
  • Python Basics with Numpy (Optional)
  • Logistic Regression with a neural network mindset
Lecture 3 09/29 Topics: Full-cycle of a Deep Learning Project (no slides) Completed modules: Quizzes (due at 8 30am PST):
  • Shallow Neural Networks
  • Key concepts on Deep Neural Networks
Programming Assignments (due at 8 30am PST):
  • Planar data classification with a hidden layer
  • Building your Deep Neural Network: step by step
  • Deep Neural Network - Application
Project Meeting #1 {{ site.course.project_timeline.proposal | date: site.course.project_timeline.syllabus_date_format }} Instructions Meet with any TA between 9/15 and 9/30 to discuss your proposal.
Project Proposal Due {{ site.course.project_timeline.proposal | date: site.course.project_timeline.syllabus_date_format }} Instructions
Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization (Course 2)
Lecture 4 10/06 Topics: Adversarial examples - GANs (slides)
  • Attacking neural networks with Adversarial Examples and Generative Adversarial Networks
Optional Readings: Explaining and Harnessing Adversarial Examples, Generative Adversarial Nets, Conditional GAN, Super-Resolution GAN, CycleGAN 		Completed modules:
  • C2M1: Practical aspects of deep learning (slides)
  • C2M2: Optimization algorithms (slides)
 Quizzes (due at 8 30am PST):
  • Practical aspects of deep learning
  • Optimization Algorithms
Programming Assignments (due at 8 30am PST):
  • Initialization
  • Regularization
  • Gradient Checking
  • Optimization
Structuring Machine Learning Projects (Course 3)
Lecture 5 10/13 Topics: AI and Healthcare. Guest Speaker: Pranav Rajpurkar. (guest slides) (main slides) Completed modules:
  • C2M3: Hyperparameter Tuning, Batch Normalization (slides)
  • C3M1: ML Strategy (1) (slides)
  • C3M2: ML Strategy (2) (slides)
 Quizzes (due at 8 30am PST):
  • Hyperparameter tuning, Batch Normalization, Programming Frameworks
  • Bird recognition in the city of Peacetopia (case study)
  • Autonomous driving (case study)
Programming Assignments (due at 8 30am PST):
  • Tensorflow
Convolutional Neural Networks (Course 4)
Midterm Review 10/20 12:00pm-1:20pm PT Past midterms:
Lecture 6 10/20 Topics: Deep Learning Strategy (no slides)

Optional Reading: A guide to convolution arithmetic for deep learning, Is the deconvolution layer the same as a convolutional layer?, Visualizing and Understanding Convolutional Networks, Deep Inside Convolutional Networks: Visualizing Image Classification Models and Saliency Maps, Understanding Neural Networks Through Deep Visualization, Learning Deep Features for Discriminative Localization 		Completed modules:
  • C4M1: Foundations of Convolutional Neural Network (slides)
  • C4M2: Deep Convolutional Models (slides)
 Quizzes (due at 8 30am PST):
  • The basics of ConvNets
  • Deep convolutional models
Programming Assignments (due at 8 30am PST):
  • Convolutional Model: step by step
  • Convolutional Model: application
  • Keras Tutorial: This assignment is optional.
  • Residual Networks
Midterm 10/21 00:00am - 10/22 11:59pm Time: 3 hours
Lecture 7 10/27 Topics: Interpretability of Neural Networks (slides) Completed modules:
  • C4M3: ConvNets Applications (1) (slides)
  • C4M4: ConvNets Applications (2) (slides)
 Quizzes (due at 8 30am PST):
  • Detection Algorithms
  • Special Applications: Face Recognition & Neural Style Transfer
Programming Assignments (due at 8 30am PST):
  • Car Detection with YOLO
  • Art Generation with Neural Style Transfer
  • Face Recognition
Project Meeting #2 {{ site.course.project_timeline.milestone | date: site.course.project_timeline.syllabus_date_format }} Instructions Meet with your assigned TA between 10/1 and 10/30 to discuss your milestone report.
Project Milestone Due {{ site.course.project_timeline.milestone | date: site.course.project_timeline.syllabus_date_format }} Instructions
Sequence Models (Course 5)
Lecture 8 11/03 Topics:
  • Career Advice
  • Reading Research Papers
Optional Reading 		Completed modules:
  • C5M1: Recurrent Neural Networks (slides)
 Quizzes (due at 8 30am PST):
  • Recurrent Neural Networks
Programming Assignments (due at 8 30am PST):
  • Building a Recurrent Neural Network - Step by Step
  • Dinosaur Land -- Character-level Language Modeling
  • Jazz improvisation with LSTM
Lecture 9 11/10 Topics: (slides)
  • Deep Reinforcement Learning

Optional Reading: 		Completed modules:
  • C5M2: Natural Language Processing and Word Embeddings (slides)
  • C5M3: Sequence-to-Sequence Models (slides)
 Quizzes (due at 8 30am PST):
  • Natural Language Processing and Word Embeddings
  • Sequence Models and Attention Mechanism
Programming Assignments (due at 8 30am PST):
  • Operations on Word Vectors - Debiasing
  • Emojify!
  • Neural Machine Translation with Attention
  • Trigger Word Detection
Lecture 10 11/17 Topics: (slides)
  • Class wrap-up
  • What's next?
 Optional:
  • If you’re interested in testing your ML/DL skills or preparing for job interviews in AI, you can take the Workera assessment
Project Meeting #3 {{ site.course.project_timeline.poster_and_report | date: site.course.project_timeline.syllabus_date_format }} Instructions Meet with your assigned TA between 11/2 and 11/17 (before class) to discuss your final project report.
Project Final Report & Video Due {{ site.course.project_timeline.poster_and_report | date: site.course.project_timeline.syllabus_date_format }} Instructions Please read over the final project guidelines here for information on the rubric and late submissions.