Company Filing History:
Years Active: 2023
Title: Vahideh Akhlaghi: Innovator in Neural Architecture Search
Introduction
Vahideh Akhlaghi is a prominent inventor based in Redmond, WA (US). She has made significant contributions to the field of artificial intelligence, particularly in the optimization of neural networks. With a total of 2 patents, her work focuses on enhancing the performance of artificial neural networks through innovative techniques.
Latest Patents
Her latest patents include "Quantization-aware neural architecture search" and "Differential bit width neural architecture search." The first patent, Quantization-aware neural architecture search ('QNAS'), is designed to learn optimal hyperparameters for configuring an artificial neural network ('ANN') that quantizes activation values and/or weights. This involves model topology parameters, quantization parameters, and hardware architecture parameters, which are crucial for defining the structure and connectivity of an ANN. The second patent, Differential bit width neural architecture search, utilizes machine learning to optimize the quantization configuration for an ANN. This includes learning the optimal bit width for quantizing weights and activation values, which can significantly improve the performance of the ANN during inference.
Career Highlights
Vahideh Akhlaghi is currently employed at Microsoft Technology Licensing, LLC, where she continues to push the boundaries of technology in her field. Her innovative work has garnered attention and respect within the industry.
Collaborations
She collaborates with notable colleagues such as Kalin Ovtcharov and Eric S Chung, contributing to a dynamic and innovative work environment.
Conclusion
Vahideh Akhlaghi is a trailblazer in the realm of neural architecture search, with her patents reflecting her commitment to advancing artificial intelligence. Her contributions are paving the way for future innovations in the field.