🔐 Homomorphic Encryption Demos

💡 Why Homomorphic Encryption?

Homomorphic Encryption (HE) refers to a new type of cutting-edge encryption technology that allows computations to be performed directly on encrypted data—without ever exposing the raw inputs. This breakthrough enables secure workflows in areas like:

• Privacy-preserving machine learning
• Federated finance and encrypted identity checks
• Secure healthcare and biometric verification

This repository showcases practical HE demos using two popular encryption schemes, BFV and CKKS, each tailored to specific use cases.

Built with:

• TenSEAL — CKKS-based encrypted vector math library
• SEAL-Python — Python wrapper for Microsoft SEAL, supporting both BFV and CKKS

🤔 Why BFV and CKKS?

This repository uses both BFV and CKKS encryption schemes because each is optimized for different types of computation:

Scheme	Precision Type	Best Use	Example Demo
BFV	Exact integers	Binary decisions, like age checks	seal_demos/
CKKS	Approximate floats	Encrypted ML inference and analytics	tenseal_demos/

• 🧮 BFV is essential when precision matters—such as verifying whether a value meets a specific threshold.
• 📊 CKKS is perfect for privacy-preserving machine learning and statistical workloads, where small rounding errors are acceptable.

📌 Note: While TenSEAL includes BFV support, its main strength and tooling focus is on CKKS. For more robust integer logic, we rely on SEAL-Python.

This split ensures each demo showcases the encryption scheme in the context it performs best—without compromise.

🗂️ Project Structure

tenseal_demos/

• Demonstrates encrypted ML inference and floating-point operations using CKKS.
• Ideal for showcasing privacy-preserving analytics.

seal_demos/

• Implements discrete logic using BFV.
• Demonstrates secure age checks without revealing user data.

---

🧪 How to Run

Each project is self-contained with its own dependencies. To get started:

1. Navigate to the appropriate folder (tenseal_demos/ or seal_demos/)
2. Follow setup instructions in the respective README.md

This modular setup ensures that library dependencies remain isolated and don’t conflict across projects.

---