Shlomit Shachor Ifergan

Title: Shlomit Shachor Ifergan: Innovator in Machine Learning Privacy

Introduction

Shlomit Shachor Ifergan is a prominent inventor based in Yokneam Eilit, Israel. She has made significant contributions to the field of machine learning, particularly in the area of privacy risk assessment. With a total of 2 patents, her work is paving the way for advancements in data security and privacy.

Latest Patents

One of her latest patents is titled "Membership inference attacks using multiple specialized machine learning models." This method involves receiving both a training dataset and a testing dataset, which are used to train and test a specific machine learning model. The process includes dividing the training dataset into non-overlapping subsets and pairing them with testing subsets to train different membership inference attack models. The goal is to evaluate sensitive information leakage from the machine learning model based on aggregated inferences by the trained models.

Another notable patent is "Assessing feature-based privacy risk in machine learning model." This invention provides a method for assessing privacy risk by identifying a dataset and a machine learning model. It includes selecting a target feature set and conducting differential inference attacks to determine a privacy risk score based on the results of these attacks.

Career Highlights

Shlomit is currently employed at International Business Machines Corporation (IBM), where she continues to innovate in the field of machine learning. Her work focuses on enhancing privacy measures and developing methods to protect sensitive information in machine learning applications.

Collaborations

Shlomit collaborates with talented individuals such as Natalia Razinkov and Abigail Goldsteen, contributing to a dynamic and innovative work environment.

Conclusion

Shlomit Shachor Ifergan is a key figure in the realm of machine learning privacy, with her patents reflecting her commitment to advancing data security. Her contributions are vital in addressing the challenges of privacy in modern technology.