Differential Privacy

Definitions

1. A mathematical framework that provides provable privacy guarantees by adding calibrated noise to query results or model training. Ensures that the output of an analysis is essentially the same whether or not any single individual's data is included. Used by Apple, Google, and the US Census Bureau.

   In plain English: Adding carefully calculated random noise to data analysis results so no individual person's data can be identified.

   Example: "The analytics pipeline uses differential privacy with epsilon=1.0 — individual user behavior can't be reverse-engineered from aggregate results."

Related Terms

• Edge AI
• Cross-Validation
• Synthetic Data
• Federated Learning
• Prompt Injection
• Conformal Prediction
• Gaussian Mixture Model
• Unlearning
• Prompt Leaking
• System Prompt Extraction
• Tor Project