Welcome to the Basics of PETs for Non-technical Practitioners! In this course, you will learn about the basics of five prominent Privacy-Enhancing Technologies. Beyond this, you will become an expert in discussing the primary benefits and limitations of each PET, as well as be familiar with some application use cases. To start off, click on the Introduction to find out about what PETs are in general, as well as why and how they can be used. We hope you enjoy this course!
Federated Learning allows to train Machine Learning models collaboratively without exchanging or centralizing raw data. This help to minimize the risk of data breaches! Click on the sections below to learn about the foundations of Federated Learning.
Differential Privacy allows for meaningful data analysis while still protecting the individuals participating in a dataset. This all comes with a mathematically grounded privacy guarantee! Click on the sections below to learn the fundamentals of Differential Privacy.
Homomorphic Encryption enables performing computations on encrypted data without the need to decrypt it first. This is useful for sensitive applications where sharing raw data is not possible. Click below to learn more about Homomorphic Encryption.
Secure Multiparty Computation allows for two or more parties to perform computation in a collaborative way, without the need for any of the individuals to share their own information. To learn more, click on the sections below.
In the case where validation is required, for example, to prove age or identity, Zero-Knowledge Proofs provide a mechanism whereby an entity can prove the requested attribute to a verifier without ever sharing the actual sensitive information itself. Learn more below!