Course Syllabus
INFO 6120 Ubiquitous Computing
Spring 2021
Tuesday & Thursday: 11:50am - 1:05pm
Instructors
Professor Tanzeem Choudhury
Email: tanzeem.choudhury@cornell.edu
Dr. Alexander Adams
Email: ata56@cornell.edu
Office Hour: Friday 9-11 AM
Location: https://cornell.zoom.us/j/95390985439?pwd=QmdiNVpnb0RPbUhGR1FCVmo0b25TQT09
TA: Yiran Zhao
Email: yz2647@cornell.edu
Office Hours: Friday 11 AM - 1 PM, or by appointment
Location: https://cornell.zoom.us/j/8017788346?pwd=anhDNVN6OVcvVGlMRjNhdkNTcWZEQT09
Overview
This course will introduce students to the field of Ubiquitous Computing – a multidisciplinary research area that draws from Signal Processing, Machine Learning, Device Making, Human Computer Interaction, as well as from Psychology and Sociology. We will highlight various challenges in data collection, representation of models, and evaluation. We will brain-storm ideas on how future research can go about tackling some of these challenges. Formal lectures, hands on exercises/lab along with discussions will be equally important aspects of the course. Students will be required to read, and critique papers and we will have a few short debates. Participation in discussions will be evaluated as well as mini projects and assignments during the term and an end of term final project.
We will normally have lecture and discussion session on Tuesdays where the instructors and students will explore a topic and discuss papers on a ubiquitous computing-relevant topic. On Thursdays there are lab sessions to work on a project related to a ubiquitous computing application. This class is focused on ubiquitous computing for health applications, but other topics will be included as well.
Prerequisites: Students should have coding proficiency and basic familiarity of sensors available on mobile devices and IoT, and a basic knowledge of machine learning. Contact the instructors if you have questions.
Course Materials
To complete the in-class labs and to prepare you for the final project, students are required to purchase a set of electronics. We generated a list that can be purchased from (mostly) AdaFruit. You could swap these parts with those with the same functions that you already have, but the lab instructions will be based on this list. The whole list should cost approximately $150.
Papers will be posted on Canvas before the course meeting.
Grading and Assignments
Your grades will be determined by assignments (30%), final project (40%), paper critiques / discussion (20%), and class participation (10%).
Assignments: 30%
There will be 3 individual assignments, each worth 10%. These assignments will involve both basic building/prototyping, as well as coding and machine learning.
Paper critiques and presentation: 20%
You are expected to read the assigned readings before class. This will typically mean two full-length research papers per week. Your are expected the read and write discussions on both readings before the start of the Tuesday's class, and post them on to the corresponding Discussion thread on canvas. Your discussion should be 1-2 pages longs, or around 500 - 750 words. You can either directly response in the Discussion, or upload a document.
However, if you choose to lead a presentation of a paper, you only need to read one paper and present it. You do not have to complete the written critique of the other reading material of that week.
Please signup for presentations here. Please use this template to guide what you should present in these presentations.
Simply criticizing the details of research often leads to an underwhelming discussion. We encourage students to draw upon their backgrounds to surface more interesting discussion topics. For some thoughts on why and how to formulate good questions, see our colleagueDan Cosley’s blog post.
You could discuss from these perspectives:
- What idea or innovation enabled this, what more might be done based on that idea or innovation? How can the ideas proposed by this research be used in the real world? What might the barriers to adoption be?
- How might this research help address gaps in other solutions or research you have seen in this space?
- How well did the authors uniquely communicate their data, ideas, and results? Could the finding have been made clearer to achieve greater impact?
- The goal of posting discussion topics is to facilitate rich discussions during class. All of your classmates will have read the paper, so do not simply post a summary of the paper.
Participation in the posting potential discussion topics for each day will be graded on a scale from 0 to 3.
- 0: If you do not participate.
- 1: If your participation seems weak and does not convince us you read, understood, and considered the readings.
- 2: If your participation shows you read and understood the readings, then surfaced a potentially interesting discussion. This will be the most common grade.
- 3: Reserved for especially strong or insightful discussion topics. This will be an uncommon grade and may not be given out for every thread/topic.
Class participation: 10%
Class participation will be determined by active in-class engagement in paper discussions, and questions or interesting observations during the lab sessions.
Final project: 40%
Your will complete a semester-long final project. You will be required to find a real-world problem, propose an innovative solution to that problem using Ubicomp technique, and demonstrate efficacy of your innovative solution or a detailed analysis of why things didn’t turn out as you expected. The project will involve both building and interacting with sensors, and are likely to involve signal processing and/or machine learning.
Submission
Submissions will be coordinated using Canvas:
Academic Integrity
You are expected to observe Cornell’s Code of Academic Integrity in all aspects of this course. The code states that:
Absolute integrity is expected of every Cornell student in all academic undertakings. Academic integrity is expected not only in formal coursework situations, but in all University relationships and interactions connected to the educational process, including the use of University resources. A Cornell student's submission of work for academic credit indicates that the work is the student's own. All outside assistance should be acknowledged, and the student's academic position truthfully reported at all times. In addition, Cornell students have a right to expect academic integrity from each of their peers.
Weekly Schedule:
-
- Week 1
- February 9, 2021 - [Lecture] Introduction and class logistics
- February 11, 2021 - [Research papers discussion] History of Ubicomp
- Weiser, M. (1991, September). The Computer for the 21st Century. Scientific American, 265(3), 94–100.
- Weiser, M., & Brown, J. S. (1997). The Coming Age of Calm Technology. In P. J. Denning & R. M. Metcalfe, Beyond Calculation (pp. 75–85).
- [Optional] Yvonne Rogers commentary on Calm Technology, ten years later. Rogers, Y. (2006). Moving on from Weiser’s Vision of Calm Computing: Engaging UbiComp Experiences. In P. Dourish & A. Friday (Eds.), UbiComp 2006: Ubiquitous Computing (Vol. 4206, pp. 404–421).
- Week 2
- February 16, 2021 - [Lecture] Intro to Electronics
- February 18, 2021 - [Lab] Intro to Electronics
- Week 3
- February 23, 2021 - [Lecture] Sensors/Sensing
- February 25, 2021 - [Lab] Sensors/Sensing
- Week 4
- March 2, 2021 - [Lecture] Signal Acquisition and Responding to Physical World
- March 4, 2021 - [Lecture] Signal Acquisition and Responding to Physical World
- Week 5
- March 9, 2021 - No Class
- March 11, 2021 - Open lab for Wearable Project
- March 12, 2021 - [Assignment 1 due]
- Week 6
- March 16, 2021 - [Lecture] Data processing and feature engineering
- March 16, 2021 - [Written final project proposal due]
- March 18, 2021- [Lab] Data processing and feature engineering
- Week 7
- March 23, 2021 - [Research papers discussion]
- The Mobile Sensing Platform: An Embedded Activity recognition System.
- SoundSense-scalable sound sensing for people-centric applications on mobile phones
- March 25, 2021 - [2-minute madness for final project]
- Week 8
- March 30, 2021 - [Lecture] Machine Learning considerations in Ubicomp
- April 1, 2021 - [Lab] Machine Learning considerations in Ubicomp
- Week 9
- April 6, 2021 – [Lab] Activity Recognition
- April 8, 2021 - [Research papers discussion] Machine Learning considerations in Ubicomp
- Deep, Convolutional, and Recurrent Models for Human Activity Recognition using Wearables
- Deep Learning in the Era of Edge Computing: Challenges and Opportunities
- [Optional] Dog's Life - Wearable Activity Recognition for Dogs
- April 8, 2021 - [Assignment 2 due]
- Week 10
- April 13, 2021 – [Research papers discussion] Biofeedback
- Calm Commute: Guided Slow Breathing for Daily Stress Management in Drivers
- BoostMeUp: Improving Cognitive Performance in the Moment by Unobtrusively Regulating Emotions with a Smartwatch
- April 15, 2021 – [Final Project Update]
- Week 11
-
-
-
- April 20, 2021 – [Research papers discussion] Unusual applications of common sensors
- April 22, 2021 – [Lab]
-
-
-
- Week 12
-
- Week 13
-
- Week 14
-
- May 11, 2021 – [Lecture] Exciting Ubicomp research from 2021
- May 13, 2021 - Final Project Presentations
- May 20, 2021 - Final Project Write-up Due
-
- Week 14