Can’t decide between parallel sessions? Here’s more information about each session to help!
Learning by Doing: Teaching Applied Differential Privacy
Date and time: Thursday, August 22nd (3:15PM-4:15PM)
![](https://sites.harvard.edu/opendp/files/2024/06/Screenshot-2024-06-21-at-9.35.16 AM-917x1024.png)
![](https://sites.harvard.edu/opendp/files/2024/06/Screenshot-2024-06-21-at-9.35.59 AM-986x1024.png)
![](https://sites.harvard.edu/opendp/files/2024/05/Screenshot-2024-05-23-at-1.06.58 PM.png)
![](https://sites.harvard.edu/opendp/files/2024/07/Salil-Vadhan.jpg)
![](https://sites.harvard.edu/opendp/files/2024/06/Cummings.png)
![](https://sites.harvard.edu/opendp/files/2024/08/Screenshot-2024-08-05-at-10.54.12 AM.png)
Speakers/Organizers: Katrina Ligett (Hebrew University of Jerusalem), Moshe Shenfeld (Hebrew University of Jerusalem), Joe Near (University of Vermont), Salil Vadhan (OpenDP/Harvard University), Rachel Cummings (Columbia University), Wanrong Zhang (TikTok)
Descriptor: This session will feature interactive demonstrations and a panel discussion with educators who have developed teaching materials for differential privacy that are oriented towards implementations and applications.
We’ll make the case that giving students hands-on experience with implementing and evaluating DP algorithms and attacks has many benefits, even for courses whose primary motivations are theoretical. And our discussion will delve into some of the challenges, quandaries, opportunities raised by teaching such a course.
Privacy Attacks & Auditing
Date and time: Thursday, August 22nd (3:15PM-4:15PM)
![](https://sites.harvard.edu/opendp/files/2024/06/Ullman-1.jpeg)
![](https://sites.harvard.edu/opendp/files/2024/06/deMontoye.jpeg)
![](https://sites.harvard.edu/opendp/files/2024/08/honaker_2.jpeg)
![](https://sites.harvard.edu/opendp/files/2024/06/Ratliff-1024x1024.jpeg)
![](https://sites.harvard.edu/opendp/files/2024/06/Sharon-Ayalde.png)
Led by: Jonathan Ullman (Northeastern University), Yves-Alexandre de Montoye (Imperial College), James Honaker (Anonym/OpenDP), Zachary Ratliff (Harvard University), Sharon Ayalde (OpenDP/Harvard University)
Descriptor: Privacy attacks and auditing are a crucial part of the pipeline of DP deployments, used for increasing awareness of privacy risks, motivating adoption of DP, stress testing proposed systems, communicating privacy guarantees, and more. Since last year’s community meeting, OpenDP has launched a working group to explore how the scientific community and OpenDP can provide accessible tools for privacy attacks and auditing. This breakout session will report on the progress made so far and gather input from the community on next steps.
Privacy Attacks Repository Sheet
This session is a continued discussion from last year’s OpenDP Community Meeting where a working group has launched and been active over the course of the year. A mix of industry, academic, and government community members have been contributing to this working group. You can find a recap of last year’s parallel breakout session here and if you’re interested in joining the working group, you can join our dedicated slack channel #wg-privacy-attacks-and-auditing!
DP Beyond Algorithms (part 1): What does it take to successfully deploy DP from a sociotechnical perspective?
Date and time: Friday, August 23rd (10:50AM-11:50AM)
![](https://sites.harvard.edu/opendp/files/2024/06/Cummings.png)
![](https://sites.harvard.edu/opendp/files/2024/06/Kaptchukis-1024x979.jpeg)
![](https://sites.harvard.edu/opendp/files/2024/06/Redmiles-683x1024.jpeg)
![](https://sites.harvard.edu/opendp/files/2024/06/Nanayakkara-1024x1024.jpg)
![](https://sites.harvard.edu/opendp/files/2024/06/Sarathy.png)
Led by: Rachel Cummings (Columbia University), Gabriel Kaptchuk (University of Maryland, College Park), Elissa Redmiles (Georgetown University), Priyanka Nanayakkara (Northwestern University), Jayshree Sarathy (Columbia University)
Descriptor: DP deployments require thinking carefully not only about algorithms, but also about sociotechnical considerations extending beyond technical tools. In this breakout session, we will discuss issues including communication around DP deployments to parties like data subjects and policymakers, participatory design, and organizational decision-making. This breakout session will form a foundation for the longer workshop on sociotechnical aspects of DP deployments.
A workshop will be held Friday, August 23rd in the afternoon to continue the conversation entitled “DP Beyond Algorithms (part 2): Building a Community around Sociotechnical Aspects of DP Deployment”. More information about this workshop can be found here!
OpenDP Research Challenges and Development Directions
Date and time: Friday, August 23rd (10:50AM-11:50AM)
![](https://sites.harvard.edu/opendp/files/2024/07/michael_shoemate_0.jpeg)
![](https://sites.harvard.edu/opendp/files/2024/05/Screenshot-2024-05-23-at-1.01.58 PM.png)
Led by: Michael Shoemate (Harvard University/OpenDP), Vikrant Singhal (Harvard University/OpenDP)
Descriptor: Expanding upon the OpenDP Technical Updates session, this parallel breakout will focus on open research problems and development contributions to the library. What are the Community’s use cases and needs that could prioritize future improvements or updates? The breakout will also share OpenDP’s roadmap and identify challenges where the Community can help.
We would appreciate anyone from OpenDP Community (whether you can make this session or not) to complete this poll to share your thoughts on prioritizing OpenDP development projects. We will use the poll results to talk through during this session so please add your responses soon!