TechPrufer
Sign inGet started

Blogs›IBM›AI Data Minimization Toolkit

AI Data Minimization Toolkit

AI Data Minimization Toolkit

1
posts
2022

The AI Privacy toolkit provides a method to reduce the amount of personal data needed for machine learning model predictions by removing or generalizing input features without significantly impacting accuracy. This helps adhere to data minimization regulations like GDPR and reduces security risks associated with large datasets. The toolkit can determine the minimal set of features and their required detail level for a given trained model. It has been applied to classification models and research is ongoing to extend it to regression and integrate with XAI and interpretability techniques. The technology is also being evaluated within the iToBoS project.

2022

Reduce data privacy issues with machine learning models

1/18/2022

Introduces IBM's open-source AI Privacy toolkit, specifically its data minimization capability for machine learning models. Details how the toolkit identifies the minimal set of features and their required detail for accurate predictions, reducing the need for extensive personal data. Explains the underlying method of removing or generalizing input features based on model knowledge. Provides a real-world application example in healthcare and mentions ongoing research and a published paper on the topic.