Glossary term
Glossary term
Safety and Alignment
In machine learning, an anonymization approach to protect any sensitive data (for example, an individual's personal information) included in a model's training set from being exposed. This approach ensures that the model doesn't learn or remember much about a specific individual. This is accomplished by sampling and adding noise during model training to obscure individual data points, mitigating the risk of exposing sensitive training data.
Differential privacy is also used outside of machine learning. For example, data scientists sometimes use differential privacy to protect individual privacy when computing product usage statistics for different demographics.
Created for this library
A health-tech startup uses differential privacy when sharing aggregated population statistics with researchers so individuals cannot be re-identified.
A bank uses differential privacy in its internal analytics platform to allow business teams to query customer data without seeing individual records.
A government statistics agency publishes census tables with differential privacy noise so small communities are protected against record linkage.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License