Sign up for the final project showcase time slots here!

Final Project Details

• Deliverables

  • A final presentation detailing your project and a demo (if applicable)

  • No reports (papers) are required, just the final presentation.

• The showcase will be held on Tuesday, Apr 23 from1:30 to 4:30 at FXB 1024 (the usual classroom)

• Attendance is mandatory for all students.

• Participation in the presentation is mandatory for all team members.

• Each team will have 15 minutes to present and 5 minutes for Q&A (a total of 20 minutes)

Final Exam Details

• You will have from 4 pm ET Tuesday, April 16 until 4 pm ET Wednesday, April 17 to work on the exam. (subjected to change)

• The exam is a take-home exam and is open- everything except AI assistance and collaboration. 

• The exam will likely take less than 30 minutes to complete.

• The exam will be on Google Forms which you'll need to fill out and submit to record your answers.

•  If you submit multiple times, we will use your final submission. 

• You can use any resource from the class (slides, presentations, videos) to prepare for the exam. However, the information necessary to complete the exam is covered in the main lectures of the class (that is, in the material presented by Prof. Austin, the students, and the guest lecturers).

• COLLABORATION or AI ASSISTANCE is NOT PERMITTED for this exam, all work must be done by you alone.

• Here is a sample exam from winter 2023. You can ignore sections we haven't covered in class.

• 4 questions from the sample exam linked above will be directly used unmodified in this exam.

Sign up for mid-term project feedback sessions here! 

Sign up for Project feedback sessions here! Two available time blocks are on Thursday, Feb 15, and Monday, Feb 19.

The deadline for project proposals has been moved to Friday, Feb 2, 11:59 PM.

Lecture Recordings are available!

Sign up for Entrees and Side dishes here!

Side Dish 🥗

• Indicate your preferences by Jan 23rd.

  • The papers on each topic will be posted as they become available

  • A list of guest speakers will also be posted soon

  • If you have a researcher or product you’d like to host, please indicate that.

Entree 🍔

• One-page Project proposal, due by Jan 30th

  • What are the goals of the project?

  • What are the mid-point milestones and results?

  • How will you evaluate the research?

• One-page Lecture Proposal, due by Jan 30th

  • Motivations for choosing this option

  • Topic to be presented

  •  Lecture outline

• The proposal must be no more than one page 

• The proposal must contain at least 250 words

Research Proposal:

• The proposal must clearly state the following:

  • What are the goals of the project?

  • What are the techniques and methodologies you will be using?

  • What are the mid-point milestones?

  • What are the deliverables (results)?

  • How will you evaluate the research?

Lecture Proposal:

• The proposal must clearly state the following:

  • What are your motivations for choosing this option?

  • What topics are going to be presented?

  • What is your Lecture outline?

  • What techniques will you be using to make the lecture interactive?

• Send your Proposal to eecs598-9-wn2024-instructors@umich.edu by Feb 2 and CC all team members.

• Port an application of interest to you to the VIP-Bench benchmark suite, so that it can be executed with encrypted secure computation

• Implement a data analysis algorithm using Homomorphic Encryption, e.g., with Microsoft SEAL

• Implement a database algorithm using sequestered encryption, e.g., with the VIP-Bench functional library

• Design and implement a secret key-sharing protocol that can create a key that no one knows, but can successfully share that key among protocol participants

• Implement a differentially private machine learning algorithm and demonstrate it is resilient to revealing individual preferences

• Devise and evaluate an efficient ORAM algorithm, and show that it reveals no information in its access patterns

• Devise and evaluate a proof-of-work algorithm that is more power-efficient than existing algorithms

• Implement and evaluate a machine learning algorithm on Intel’s SGX TEE

• Analyze the privacy capability provided by existing browsers, and define metrics to compare and contrast

• Design and evaluate a privacy-aware recommendation system

• Explore hardware-based sensitive data sequestration for IoT systems, which would be hardware lockboxes that allow sensitive data to be always encrypted, but still allow for some processing on the protected data

• Develop a technique for secure data deletion that targets newer storage technologies, such as RRAM, NAND-Flash, etc.

• Perform a latitudinal study of the privacy of a particular application class, e.g., social media apps, self-help apps, etc.

• Explore side channels that may exist in PET protocols, such as FHE and MPC

• Explore privacy-enhanced front ends to traditionally privacy-adverse applications, such as search engines, social media platforms, etc.