Week 1: Course Introduction

Announcements

Wednesday

Introduction to CS91.3

Welcome to CS91.3! This course is designed to have a mix of lecture and hands-on, in-class exercises. This page will review some of the tools we will use during the labs and help you get set up with the system in the CS department and your personal computer.

We value all student feedback. Please be constructive in any comments so that we can adjust the course as best possible. This semester, we are using EdSTEM to manage course discussions and announcements.

Who should take this course?

This course is designed for computer science students interested in Machine Learning (ML) and Brain-Computer Interfaces (BCI). Previous ML and BCI experience are nice to have but NOT required. Previous research experience is nice to have but NOT required.

Adding the course, switching sections

If you like to add this course or switch to a different lab section, please contact your course instructor. If you are already on the waitlist, please email your course instructor to sign in at the beginning of each class.

Course Webpage

The Course Website is updated regularly and contains recent course announcements, links to in-class exercises, lab assignments, quiz topics, and other less dynamic course information, including office hours, possible TA sessions, and research tips. Please Bookmark the website in Firefox or Chrome and refer to it regularly. Thanks.

Please be aware that many elements on the course website will change throughout the semester, including the course schedule. It is your responsibility to review the course website periodically for updates.

EdSTEM

This semester, we use EdSTEM to manage course discussions and announcements. If you are registered for the course, you should be automatically enrolled on EdSTEM. If you’re unable to log in or access the course discussion forum, please get in touch with your course instructor — it’s possible something may not be configured correctly for you, especially if you joined after the initial registration phase.

Labs

This course has a mandatory lab section. Attendance is required unless you have completed and submitted the lab for the week before the start of the lab. Labs usually start Wednesday and are due the following Tuesday night.

What is Research?

Three key elements, Innovation, Connection, and Impact.

*Research definition
*Research problems
*Research design
*Research analysis
*Research results
*Research papers
*Other related research topics

Innovation, create or discover something new, an idea, or a method not existing.

Connection, what other researchers have done on this topic? Is there a research gap?

Impact, is the problem important? Is the result significant? Potential to impact how many people? For how long?

Research Paper examples:

Chick the title of each paper. You should see the paper details page.

Please pay attention to the Paper PDF files next to the paper title, and Reviewers' Comments

Let’s start with these two papers, from the section of Datasets & Benchmarks Best Paper Awards

Reduced, Reused and Recycled: The Life of a Dataset in Machine Learning Research By Bernard Koch, Emily Denton, Alex Hanna, and Jacob Gates Foster.

ATOM3D: Tasks on Molecules in Three Dimensions By Raphael John Lamarre Townshend, Martin Vögele, Patricia Adriana Suriana, Alexander Derry, Alexander Powers, Yianni Laloudakis, Sidhika Balachandar, Bowen Jing, Brandon M. Anderson, Stephan Eismann, Risi Kondor, Russ Altman, and Ron O. Dror.

Here’s a brief introduction about how to review a research paper. Reviewer Guide

Please read the following two sections for this week: Review content, and Examples of Review Content.

Flipped Classroom Approach:

Please review the videos:

Background about this video and the slides: New in Machine Learning Research Video Link: Video Link, please click

Invited Talk: Sinead Williamson (UT Austin, ICML Newcomers Chair)

start from 4 hours 05 minute, end at 4 hours 40 minute of this video

Invited Talk: Nicolas Le Roux (Google Brain Montreal)

start from 3 hours 20 minute, end at 4 hours 05 minute of this video

In-class exercises:
*Explain to your classmates what's in these two videos and your understanding of research.
*Explain to your classmates which part of the video you find most helpful and why.
*Example: start from 4 hours 25 minutes, end at 4 hours 40 minutes, 'how to find research mentors'.
Social research methods

Research Methods Knowledge Base For this week, focus on the table of content first.

More informal, but could be helpful during brainstorming ideas
In-class exercises:
*Explain to your classmates what's research, with an example.

What is CS Research?

Let’s take a look at examples: Top CS Publications

In-class exercises:

*Explain to your classmates the popular topics in these CS research publications.
*Explain to your classmates what topic(s) you may be interested in and why.

What is Machine Learning (ML)?

What’s the relationship between Machine Learning, Deep Learning, and Artificial Intelligence? Let’s take a look at examples: Top AI Publications

NeurIPS, our ten-page paper, final project is based on this.

AAAI, our two-page poster, mid-term project is based on this.

In-class exercises:

*Explain to your classmates what is Machine Learning
*Explain to your classmates the popular topics in AAAI 21 student posters.

Friday

Today’s Topics

  • Brain-Computer Interfaces (BCI)

  • Interdisciplinary Research

  • Our Research in ML and BCI

  • Reading Papers and Reviews, from NeurIPS best papers

  • Paper 1, Reduced, By Bernard Koch, etc.

  • Paper 2, ATOM3D, By Raphael John Lamarre Townshend, etc.

  • Paper 3, EEGEyeNet, by Ard Kastrati, etc.

Brain-Computer Interfaces (BCI)

What is BCI? What is EEG?

In-class exercises:

*Explain to your classmates what is BCI?
*Explain to your classmates what are non-invasive BCI headsets used for?

Interdisciplinary Research

What is it? NSF NSF Support

Our Research in ML and BCI

Start with reading these three papers together.

Lab 1

  • Read the three papers (ten-page) and reviews from NeurIPS mentioned in class.

  • Learn how to write a review for a paper.

  • Deliverable for this week, review this poster, KnowBias (two-page) with the format mentioned in NeurIPS reviewer guide.

  • Lab 1 out Wednesday 01/19, due Tuesday 01/25, please email me your Lab 1 write-up as a PDF file.

  • This lab 1 is an individual assignments.

  • There will be a step-by-step guide page for each lab, similar to CS21.

  • Lab is usually out on Wednesdays, and due the following Tuesdays before midnight.