0% found this document useful (0 votes)
14 views29 pages

CS230 Deep Learning Course Overview

CS230: Deep Learning is a course that covers various topics in deep learning, including neural networks and machine learning strategies, with a focus on practical coding assignments and projects. Students are expected to complete weekly modules, quizzes, and programming assignments, with a grading formula that includes attendance, quizzes, assignments, midterms, and final projects. The course also emphasizes collaboration, mentorship, and the importance of having a solid foundation in computer science, probability, and linear algebra.

Uploaded by

rellabelas
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
14 views29 pages

CS230 Deep Learning Course Overview

CS230: Deep Learning is a course that covers various topics in deep learning, including neural networks and machine learning strategies, with a focus on practical coding assignments and projects. Students are expected to complete weekly modules, quizzes, and programming assignments, with a grading formula that includes attendance, quizzes, assignments, midterms, and final projects. The course also emphasizes collaboration, mentorship, and the importance of having a solid foundation in computer science, probability, and linear algebra.

Uploaded by

rellabelas
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

CS230: Deep Learning

Kian Katanforoosh
Today’s outline

I. Course logistics
II. Overview of CS230 programming assignments

III. Examples of student projects

Kian Katanforoosh
Course Logistics

5 “courses”:
C1: Neural Networks and Deep Learning
C2: Improving Deep Neural Networks
C3: Strategy for Machine Learning Projects
C4: Convolutional Neural Networks
C5: Sequence Models

Example: C2M3: Course 2 Module 3

The schedule is on [Link]

We use Coursera: [Link]

The class forum is on Piazza: [Link]/stanford/winter2021/cs230


Kian Katanforoosh
One week in the life of a CS230 student

1 module

Complete
Watch videos on Coursera Solve quiz
programming assignments
≈1h ≈20min
≈1-2h

15min project
Zoom lecture TA sections
1 week of CS230 ≈ 2 modules + (every Tuesday) + (every Friday) + mentorship
w/ TA
≈1h20 ≈ 1 hour
(every ≈2 weeks)

Assignments and Quizzes are due every Thursday at 8.30am PDT

Kian Katanforoosh
Grading Formula

Grade = 0.02A + 0.08Q + 0.25P a + 0.25M + 0.40P r

A = Attendance
Q = Quizzes
Pa = (Programming) assignments
M = Midterm
Pr = Final-project

Active Piazza participation = 1% bonus

Kian Katanforoosh
Late days

Example: For next Thursday at 8.30am you have


to complete the following assignments:
- 2 Quizzes:
★ Introduction to deep learning
★ Neural Network Basics

- 2 Programming assignments:
★ Python Basics with Numpy
★ Logistic Regression with a neural network mindset

At 7am on Thursday: you submit 1 quiz and the 1 PA.


At 3pm on Thursday: you submit the second quiz.
At 2pm on Friday: you submit the second PA.

How many late days did you use? 3 late days


Kian Katanforoosh
(Soft) Prerequisites

Students are expected to have the following background, and are invited to take
the Workera technical assessments prior to the class to self-assess themselves prior to
taking the class:
- Knowledge of basic computer science principles and skills, at a level sufficient to
write a reasonably non-trivial computer program. This corresponds to a Developing
level (or badge) in the “Algorithmic Coding” section on Workera.
- Familiarity with the probability theory (CS 109 or STATS 116), which students can
assess by taking the “Data Science” section on Workera.
- Familiarity with linear algebra (MATH 51), which students can assess by taking the
“Mathematics” section on Workera.

(Recommended) Take the Workera assessment ([Link]) prior to starting the


class, you’ll take it again at the end of the class to measure your progress!

Kian Katanforoosh
Today’s outline

I. Course logistics
II. Overview of CS230 programming assignments

III. Examples of student projects

Kian Katanforoosh
Projects: SIGN language image classification

y=0 y=1 y=2 y=3 y=4 y=5

⎡1 ⎤ ⎡0 ⎤ ⎡0 ⎤ ⎡0 ⎤ ⎡0 ⎤ ⎡0 ⎤
⎢0 ⎥ ⎢1 ⎥ ⎢0 ⎥ ⎢0 ⎥ ⎢0 ⎥ ⎢0 ⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢0 ⎥ ⎢0 ⎥ ⎢1 ⎥ ⎢0 ⎥ ⎢0 ⎥ ⎢0 ⎥
⎢0 ⎥ ⎢0 ⎥ ⎢0 ⎥ ⎢1 ⎥ ⎢0 ⎥ ⎢0 ⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢0 ⎥ ⎢0 ⎥ ⎢0 ⎥ ⎢0 ⎥ ⎢1 ⎥ ⎢0 ⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎣0 ⎦ ⎣0 ⎦ ⎣0 ⎦ ⎣0 ⎦ ⎣0 ⎦ ⎣1 ⎦

Kian Katanforoosh
Projects: others
85% Jeff

Optimal goalkeeper shoot


Car detection Face recognition
prediction

Music generation Text generation


Art generation

“I love you”

Emojifier Machine translation Trigger word detection

And many more…


Kian Katanforoosh
Assignment: Car detection for autonomous driving

[Deep Learning Specialization] Kian Katanforoosh


Projects: others
85% Jeff

Optimal goalkeeper shoot


Car detection Face recognition
prediction

Music generation Text generation


Art generation

“I love you”

Emojifier Machine translation Trigger word detection

And many more…


Kian Katanforoosh
[L. Gatys et al.: Image Style Transfer Using Convolutional Neural Networks , 2015] Kian Katanforoosh
[L. Gatys et al.: Image Style Transfer Using Convolutional Neural Networks , 2015] Kian Katanforoosh
[L. Gatys et al.: Image Style Transfer Using Convolutional Neural Networks , 2015] Kian Katanforoosh
Projects: others
85% Jeff

Optimal goalkeeper shoot


Car detection Face recognition
prediction

Music generation Text generation


Art generation

“I love you”

Emojifier Machine translation Trigger word detection

And many more…


Kian Katanforoosh
Today’s outline

I. Course logistics

II. Overview of CS230 programming assignments

III. Examples of student projects

Kian Katanforoosh
Projects: others

LeafNet: A Deep Learning Solution to Tree Species Identification

Predicting price of an object from a picture

Neural
300$
Network

[Galbally, Rao & Pacalin: Spring 2018, [Link]


[Steven Chen: Fall 2017] Kian Katanforoosh
Projects: others

Detect cards from real-time video of tournaments to improve


viewer understanding and accessibility

[Culberg: Winter 2019, [Link] Kian Katanforoosh


Projects: others

font-gen: Deep Models for Inferring Alternate Language Sets


from Fonts

[Fernandez: Fall 2019, [Link] Kian Katanforoosh


Projects: others

The AI project development lifecycle

[The AI project development lifecycle, Workera (2020): [Link]] Kian Katanforoosh


Projects: others

What technical roles make up an AI team

[The AI project development lifecycle, Workera (2020): [Link]] Kian Katanforoosh


Projects: others

NBA 2k19 DeepBaller: A NN-Controlled Real-Time video game AI

[Pontius, Sakata and Santos: Spring 2019, [Link] Kian Katanforoosh


Projects: others

Image-to-Image translation with Conditional-GAN

[Hu, Yu & Yu, Spring 2018: [Link] Kian Katanforoosh


Projects: others

Discrete reasoning in natural language processing

[Albrecht, Wang & Zhu: Spring 2019, [Link] Kian Katanforoosh


AI+X: Leveraging your subject-matter expertise

• Roy, Quill, and Tuchman from Material Science & Engineering predicted the
melting point and viscosity of ionic liquids based on the component anion and
cation chemical structures (report poster).
• Buechler from Mechanical Engineering developed a deep learning framework
to approximate the outputs from a power flow simulation, and evaluate
performance for a variety of power network characteristics (report poster).
• Sokol and Aguirre from the Biomedical Informatics Training Program
leveraged deep learning to estimate the ancestral composition of a genomic
sequence at high resolution (report poster).
• Peng, Zhao, Yu from Computer Science, Civil Engineering, and Biomedical
Engineering used deep learning to classify gestures from divers
communicating with an autonomous robot companion in dangerous underwater
environments (report poster).
• O’Day, Seagers, and Lee from Bioengineering and Mechanical Engineering
studied neural signals of patients with Parkinson’s disease while walking to
predict freezing behaviors (report poster).

Kian Katanforoosh
And many more…

Predicting atom energy based on atomic-structure


Visual Question Answering
Cancer/Parkinson/Alzheimer detection
Activity recognition in video
Music genre classification / Music Compression
Accent transfer in a speech
Generating images based on a given legend
Detecting earthquake precursor signals

Check out past projects: [Link]

Kian Katanforoosh
To sum up

1. You will learn about wide range of deep learning topics


2. The course is very applied, you will code these applications
3. You have access to mentorship to build an outstanding project in 10 weeks
For next Thursday (01/21) 8.30am:
- Create Coursera account and join the private session using the invitation
- Finish C1M1 & C1M2
- 2 Quizzes:
★ Introduction to deep learning
★ Neural Network Basics
- 2 Programming assignments:
★ Python Basics with Numpy
★ Logistic Regression with a neural network mindset
- Find project team-mates and fill-in the Google form that will be posted on Piazza.
- The AWS GPU credit registration form will be posted with instructions on Piazza before
class. Upon forming a team, students should sign up for GPU credits no later than
Wednesday, 1/20, at 11 PM PST.
- Students can use the “Search for Teammates!” option on Piazza to look for teammates as
we are all remote and this can prove cumbersome as compared to normal times.

This Friday (01/15):


- TA section “Getting Started with Your Project”

Download your iPython notebooks after you finished them!


You can find all deadlines on the website Syllabus Kian Katanforoosh

You might also like