Fatemeh Sheikholeslami

Title: Fatemeh Sheikholeslami: Innovator in Machine Learning

Introduction

Fatemeh Sheikholeslami is a prominent inventor based in Pittsburgh, PA (US). She has made significant contributions to the field of machine learning, holding a total of 4 patents. Her work focuses on enhancing the performance of machine learning models through innovative methods and systems.

Latest Patents

One of her latest patents is titled "Data augmentation for domain generalization." This patent discloses methods and systems for generating training data for a machine learning model to improve its performance. The process involves selecting a source image from an image database and a target image, utilizing an image segmenter to create segmentation masks for both images. By determining the foreground and background regions, the target image's foreground is replaced with the source image's foreground, resulting in an augmented image that enhances the training data for the model.

Another notable patent is "Method and system for probably robust classification with detection of adversarial examples." This computer-implemented method trains a machine-learning network by receiving input data from a sensor, which may include perturbations indicative of image, radar, sonar, or sound information. The method aims to obtain a worst-case bound on classification error and loss for perturbed input data, ultimately training a classifier that can detect an additional abstain class.

Career Highlights

Fatemeh has worked with esteemed organizations such as Robert Bosch GmbH and Carnegie Mellon University. Her experience in these institutions has allowed her to develop and refine her innovative ideas in machine learning.

Collaborations

Some of her notable coworkers include Jeremy Zico Kolter and Filipe J Cabrita Condessa. Their collaboration has contributed to the advancement of research in the field.

Conclusion

Fatemeh Sheikholeslami is a trailblazer in the realm of machine learning, with her patents reflecting her commitment to innovation and excellence. Her work continues to influence the development of more robust and effective machine learning models.