Scribd
Upload
257 views7 pages

Standford University, Course Links To "Practical Machine Learning" (2021 Fall)

This document outlines the goals, structure, assignments, exams, and projects for the Stanford CS 329P Practical Machine Learning course offered in Fall 2021. The goals of the course are to cover practical ML topics often omitted from other courses like data collection/preprocessing, model training techniques, and deployment considerations. The course will include homework assignments, a midterm exam, and a semester-long team project applying ML techniques to a practical research problem. Students will complete assignments in Python and presentations/reports in ICML conference paper style.

Uploaded by

jgalvezf
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
257 views7 pages

Standford University, Course Links To "Practical Machine Learning" (2021 Fall)

This document outlines the goals, structure, assignments, exams, and projects for the Stanford CS 329P Practical Machine Learning course offered in Fall 2021. The goals of the course are to cover practical ML topics often omitted from other courses like data collection/preprocessing, model training techniques, and deployment considerations. The course will include homework assignments, a midterm exam, and a semester-long team project applying ML techniques to a practical research problem. Students will complete assignments in Python and presentations/reports in ICML conference paper style.

Uploaded by

jgalvezf
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
You are on page 1/ 7

CS 329P : Practical Machine Learning (2021 Fall)

0. Logistics

Qingqing Huang, Mu Li, Alex Smola

https://c.d2l.ai/stanford-cs329p
Goals

• Machine Learning topics that matter but are often skipped


• Data: collection, preprocessing, features, dataset shift, non-IID data
• Model training: model selection, model tuning, transfer learning,
distillation, multimodality, scalability
• Deployment: efficiency, serving, fairness
• This is a practical course: write code (in Python)
• Project: practical ML research in a team (this way you’ll use the
techniques for real)

Stanford CS 329P (2021 Fall) - https://c.d2l.ai/stanford-cs329p


Course

• Time: Wed and Fri 9:45 -11:15 AM


• Slides and notebooks at https://c.d2l.ai/stanford-cs329p
• Discussion: piazza.com/stanford/fall2021/cs329p code: p932sc
• Videos on Youtube in 1-2 weeks after class
(we need to record without students to comply with FERPA)
• No textbook (d2l.ai has some info), more pointers in slides
• Prerequisites: Python, basic statistics and machine learning

Stanford CS 329P (2021 Fall) - https://c.d2l.ai/stanford-cs329p


Homework

• 4 assignments (each homework is 10% of the course)


• Due in 10/6, 10/20, 11/3, and 11/12
• Will be posted 2 weeks before the deadline
• Reference solutions may be based on students’ submissions
(we will acknowledge the student if we do so)
• You can be late by a total of 2 days for any reason
• 50/50 mix of programming and theory
• Team work (maximum group size is 4)

Stanford CS 329P (2021 Fall) - https://c.d2l.ai/stanford-cs329p


Exam

• Midterm exam only (10% of the course but must not fail)
• 10/27
• Open book but not open computer/phone/tablet etc …
• Problems will be similar to homework
• Why
• At least one individual test in person (so you can’t outsource the class)
• No final exam due to projects
• Helps us to understand what we should change (if anything)

Stanford CS 329P (2021 Fall) - https://c.d2l.ai/stanford-cs329p


Project

• Centerpiece of the class (50% of the course).


• Register a team and a tentative project by October 1, 2021 (10%)
• Midterm presentation on October 15, 2021 (20%)
You must produce 1 pages documentation. ICML style files.
• Review project with the TAs by November 3, 2021 (10%)
TA sign-off counts for 10% of the course.
• Project presentation on December 8 and 10, 2021 (20%)
• Written project report counts 40%.
6 pages (up to 10 more for references & appendix). ICML style files.
This is a teamwork effort.

Stanford CS 329P (2021 Fall) - https://c.d2l.ai/stanford-cs329p


Have fun, learn stuff

Stanford CS 329P (2021 Fall) - https://c.d2l.ai/stanford-cs329p

You might also like