Introduction

In this software project, we will be building a "Privacy Evaluator" for machine learning models.

The first class takes place on Tuesday, 13th of April. We always start at 8:30 and have class until 10.

Classroom

Team 1

(Every second Wednesday, starting from April 21st, 12:00-14:00)

--> https://fu-berlin.webex.com/fu-berlin/j.php?MTID=m08efac92580ffa62f05024b0b2875ee2

Members: Milos, Claas, David, Jakob, Anna, Friedrich, Marisa, Juri

Team 2

(Every second Thursday, starting from April 29th, 16:00-18:00)

--> https://fu-berlin.webex.com/fu-berlin/j.php?MTID=m8714e80dd620147409d737c724f99263

Members: Ina, Yuxuan, Janis, Florian, Tobias, Henrik, Marie

Tools

For communication in the project, we'll use the following additional tools:

- Exchange on the project: Gitter (https://gitter.im/privML/community#)

- Code and Project Backlog: Github (https://github.com/privML)

 

Preparation

Don't forget to read the literature indicated below. You will need the knowledge in order to implement the privacy evaluator.

Also familiarize yourself with Scrum. We'll be using it in the project, and therefore, every participant needs to know how it works. We will do a graded quiz about Scrum in the first week and only students who pass the quiz will be allowed to proceed in the project.

 

Literatur

 

[1] Hunt, Tyler, Congzheng Song, Reza Shokri, Vitaly Shmatikov, and Emmett Witchel. "Chiron: Privacy-preserving machine learning as a service." arXiv preprint arXiv:1803.05961 Add to Citavi project by ArXiv ID (2018)

[2] Shokri, Reza, Marco Stronati, Congzheng Song, and Vitaly Shmatikov. "Membership inference attacks against machine learning models." In 2017 IEEE Symposium on Security and Privacy (SP), pp. 3-18. IEEE, 2017

[3]  Fredrikson, Matt, Somesh Jha, and Thomas Ristenpart. "Model inversion attacks that exploit confidence information and basic countermeasures." In Proceedings of the 22nd ACM SIGSAC Conference on Computer and Communications Security, pp. 1322-1333, 2015

[4] https://www.scrumguides.org/download.html