Yifat Schacter

Title: Yifat Schacter: Innovator in Classification Techniques

Introduction

Yifat Schacter is a notable inventor based in Tel Aviv, Israel. She has made significant contributions to the field of classification techniques, holding 2 patents that showcase her innovative approach to data analysis and machine learning.

Latest Patents

One of her latest patents is titled "Just in Time Classifier Training." This patent discloses a system and method that can be utilized with any underlying classification technique. The method receives a test dataset and identifies the features present in that dataset. It then modifies the training dataset to include only those features, allowing for a tailored calibration of the classifier for incoming data. This process is repeated for each new dataset, ensuring timely adjustments to the classifier.

Another significant patent is "Reasoning Classification Based on Feature Perturbation." This invention evaluates the impact of perturbing each feature by bootstrapping it with negative samples. By measuring changes in the classifier output, users can assess the importance of specific feature values in the classified feature vector. This method enhances the understanding of how each feature influences classification outcomes.

Career Highlights

Yifat Schacter is currently employed at Microsoft Technology Licensing, LLC, where she continues to develop innovative solutions in the realm of classification techniques. Her work is instrumental in advancing the capabilities of machine learning applications.

Collaborations

Throughout her career, Yifat has collaborated with notable colleagues, including Hanan Shteingart and Yair Tor. These partnerships have contributed to her success and the development of her patented technologies.

Conclusion

Yifat Schacter is a pioneering inventor whose work in classification techniques has the potential to transform data analysis. Her patents reflect her commitment to innovation and her ability to address complex challenges in machine learning.