An End-to-end Framework for Privacy Risk Assessment of AI Models
- Abigail Goldsteen
- Shlomit Shachor
- et al.
- SYSTOR 2022
Since joining IBM in 2007, I have worked on several projects in the area of data security and privacy and data classification.
I also contributed for a short while to a Clinical Genomics decision support system.
My current research area is around AI privacy, researching privacy risks related to ML models and developing mechanisms for preserving the privacy and compliance of ML models. This includes model anonymization, data minimzation and the right to be forgotten from ML models.