This is a colocated EECS221/NETSYS270-class. Class capacity increased from 45 to 60.
|Lectures:||Tue, Thu, 2-3:30pm, SSTR 103|
|Office Hours:||Tue 12:30-1:45pm, EH 4207|
|Reading:||No textbook. Please see Reading List and Schedule.|
|Prerequisites:||None other than willingness to do independent study at PhD level. Familiarity with either systems (mobile devices, computer networks) and/or analysis (algorithms, data science) would be a plus when picking a project.|
We are moving rapidly towards a highly connected, data-rich world, where people and spaces are continuously monitored and controlled via mobile and IoT devices. Smartphones, with their seamless connectivity and access to sensors and personal data, leave rich digital traces of user activity in the physical and online world. Likewise, smart homes, offices and public spaces are increasingly equipped with IoT devices that communicate among themselves, the cloud and their operators, and they sense/actuate various aspects of the environment. On the positive side, data collected from mobile and IoT devices provide utility: a wide range of services for individuals, value for private companies, and benefits for communities, cities, and the society as a whole. On the downside, collecting sensitive data from these devices and sharing them among different entities poses significant privacy and data transparency challenges that can only be addressed through a combination of technical and policy solutions.
This course is intended for PhD (or advanced MS) students, who are interested in doing research in the area of Privacy/Data Transparency of Mobile and IoT Devices.
The first goal of the course is to expose the graduate students to selected advanced topics in this area, including basic concepts in privacy and data transparency and their applications to real-world mobile/IoT devices and networks. In particular, we will cover the following topics:
- Differential Privacy: book, paper (and for non-technical audiences); Simons Bootcamp (mostly on DP) and tutorials therein;
- Contextual Integrity by H. Nissenbaum: book, paper, presentation
- Data Transparency
- Distributed Learning and Privacy:
- Adversarial Machine Learning:
- Fairness notions
- Locational Privacy: location [EFF], Trajectory, Matchability; NYT article.
- Telephone: metadata
- Mobile Tracking and Advertising. [cross device]
- Web Tracking: the Princeton project, 3rd party tracking survey, 1-million-site-tracking, WSJ article
- IoT Privacy: IoT Inspector, Amazon Echo, Amazon Ring, "the house that spied on me", on IoT Privacy by EPIC, FTC Staff Report 2015, smart TV tracking, smart TV vulnerabilities
- IoT S&P Reading list
- Fingerprinting and personalization
- Data Transparency Lab and DTL Tools
- Search Engines and Recommendations: de-anonymizing the NetFlix Dataset
- Social Networks: graph de-anonymization, percolation matching, SecGraph (tools, paper)
- Misc: seeing through WiFi , NIST synthetic data challenge
- TOR: [design] and [CCS'13]
- EU's GDPR (General Data Protection Regulation), summary, GDPR as poetry :-)
- California Consumer Privacy Act
- Technical vs Legal Notions of Privacy
- New Deal on Data
- Topics on Human Subjects Research
- IoT Privacy landscape: [article by ACLU].
- Book on Internet Privacy Law
The second goal of the course is to train the students in carrying out research: being able to identify a problem and carry out research in a timely manner; reviewing related literature and reading paper in a critical way; communicating the output of your research through presentations and technical papers. In this course, students will get familiar with this process (i) by presenting, discussing and reviewing papers in the aforementioned areas and (ii) by doing one research project that will lead to a workshop-quality paper.
Deliverables and Grading
|In class||Paper Presentations||20%||depending on the # of students|
|Quizzes/Homeworks||20%||every 2-3 weeks|
|Project||Project Presentation||30 %||last week of quarter in class|
|Project Writeup||30 %||by final time, instead of exam|
In class: In every class meeting, we will have typically one main paper assigned, maybe two if we want to contrast them; and a few related (optional) ones. You will be expected to come to class having already read the main paper. One student will present and critic the paper for 10-15 minutes. The rest of the students are expected to participate in the discussion. Occasionally there will be an in-class quiz on the paper of the day; occasionally the quiz will be allowed to be completed home (as a homework). This process will take approximately half of the time. The remaining will be a lecture by the instructor.
Projects: The main deliverable in this course will be a research project, equivalent to a workshop paper. The instructor will propose a list of projects but you can also suggest your own. You are supposed to complete your project in groups of two. You will present your results in the last week of classes, and you will submit your writeup during final week. There is NO midterm of final exam. Ideally, your paper should be of publishable quality, in which case, I will help you submit it and will sponsor your expenses for presenting it in a conference. Project milestones:
|When||What To Do about the Project|
|immediately||form your team, browse papers, talk to instructor, define your project|
|end of 4th week||Project proposal (5min) presented in class + Proposal (ppt) due|
|last week of quarter||Final project presentation in class|
|end of finals week||Final project report (workshop paper format - 6pp)|
|throughout the quarter||talk to your instructor for feedback|
The syllabus page shows a table-oriented view of the course schedule, and the basics of course grading. You can add any other comments, notes, or thoughts you have about the course structure, course policies or anything else.
To add some comments, click the "Edit" link at the top.