I'm taking the #30DaysOfFLCode Challenge

🌟 I'm taking on the #30DaysOfFLCode Challenge to learn more about #SMPC (Secure Multi-Party Computation) and Privacy Enhancing Technologies over the next month! Join me on this journey! 📅🚀

👉 Want to learn more about #PETs? Check out 30daysofflcode.com for resources and insights!

The method used by Alice and Bob to find the average salary is called additive secret sharing, but there are many other protocols available for different scenarios. 📚🔍

In model inference, SMPC lets clients request data from servers while keeping both sides' info private.

During training, groups can combine datasets without revealing sensitive information—perfect for Federated Learning, where models are trained collaboratively but privately! 👥📊🌍

The power of MPC extends beyond simple calculations like salary averages; it's crucial for privacy-preserving applications across various fields—especially in Machine Learning (ML)! 🤖💻

Step 3: They sum up the numbers they currently have and share those sums with each other.

Finally, averaging these sums gives them the average salary without exposing any exact figures! 📊🤫

Step 1: Alice randomly shares one of her two numbers with Bob.

Step 2: Bob does the same, sharing one of his numbers back with Alice. Now both have two numbers each! 🔄🤝

Here’s how it works in our example:
- Both Alice and Bob generate two random numbers each that add up to their salaries.
- These numbers don't reveal any individual salary until summed up! 🔢💡

What is SMPC?🤔
Secure Multiparty Computation allows multiple parties to compute a function over their inputs while keeping those inputs private. It’s like magic for privacy! 🎩✨

Imagine Alice and Bob want to know their average salary. Normally, they’d need to trust a third party with their sensitive information. But what if that trusted entity mishandles their data? 😟 This is where SMPC shines! 🤝🔒