Ritwika Chowdhury

Title: Ritwika Chowdhury: Innovator in Neural Network Acceleration

Introduction

Ritwika Chowdhury is a prominent inventor based in Bengaluru, India. She has made significant contributions to the field of machine learning, particularly in the acceleration of neural networks. Her innovative approach focuses on enhancing computational efficiency through low precision-based multiplication techniques.

Latest Patents

Ritwika holds a patent titled "Accelerating neural networks with low precision-based multiplication and exploiting sparsity in higher order bits." This invention discloses an apparatus designed to facilitate the acceleration of neural networks by utilizing low precision-based multiplication and exploiting sparsity in higher order bits. The apparatus includes a processor with a re-encoder that re-encodes signed input numbers from a higher precision format to a lower precision format. Additionally, it features a multiply-add circuit that performs operations in the higher precision format using the re-encoded numbers, along with a sparsity hardware circuit that reduces computation on zero values.

Career Highlights

Ritwika is currently employed at Intel Corporation, where she continues to push the boundaries of technology in her field. Her work is instrumental in advancing the capabilities of machine learning models, making them more efficient and effective.

Collaborations

Ritwika collaborates with talented individuals such as Avishaii Abuhatzera and Om Ji Omer, contributing to a dynamic and innovative work environment.

Conclusion

Ritwika Chowdhury is a trailblazer in the realm of neural network acceleration, with her patent showcasing her commitment to innovation. Her contributions are paving the way for more efficient machine learning technologies.