Pittsburgh, PA, United States of America

Diana Marculescu

This inventor holds 1 USPTO granted patent and 1 published patent application. Top assignee: Carnegie Mellon University. Active years: 2022.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Diana Marculescu. Retrieved from https://idiyas.com/inventor/diana-marculescu

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile

USPTO Granted Patents = 1 

% Patents Active = 100.0

Average Co-Inventor Count = 5.0

ph-index = 1


Company Filing History:


Years Active: 2022

Loading Chart...
1 patent (USPTO):Explore Patents

Title: Diana Marculescu: Innovator in Deep Neural Networks

Introduction

Diana Marculescu is a prominent inventor based in Pittsburgh, PA (US). She has made significant contributions to the field of deep learning and neural networks. Her innovative work focuses on improving the efficiency and performance of deep neural networks (DNNs) through novel approaches.

Latest Patents

Diana Marculescu holds a patent for "Flexible, lightweight quantized deep neural networks." This invention aims to enhance the throughput and energy efficiency of DNNs on customized hardware. By constraining the weights of DNNs to a limited combination of powers of 2, her approach allows for the replacement of the multiply-accumulate operation with a single shift operation or two shifts and an add operation. Furthermore, the invention provides design flexibility by allowing the optimal selection of k for each convolutional filter, rather than fixing it for every filter. This differentiable selection process enables model training to determine k-based weights on a per-filter basis. The invention achieves higher speeds compared to traditional lightweight neural networks with minimal accuracy degradation, while also enhancing computational energy efficiency for ASIC implementation. She has 1 patent to her name.

Career Highlights

Diana Marculescu is affiliated with Carnegie Mellon University, where she continues to advance her research in neural networks and machine learning. Her work has garnered attention for its practical applications and innovative methodologies.

Collaborations

Diana has collaborated with notable colleagues, including Ruizhou Ding and Zeye Liu. Their joint efforts contribute to the ongoing research and development in the field of deep learning.

Conclusion

Diana Marculescu is a trailblazer in the realm of deep neural networks, with her patented innovations paving the way for more efficient and effective machine learning applications. Her contributions are shaping the future of technology and research in this critical area.

Profile summary based on public USPTO records.
Data Sources: USPTO Patent Grant XML, Patent Center, EPO & CIPO • Normalized by IDiyas Innovation Graph. Methodology & provenance architecturePlease report any incorrect information to [email protected]
Loading…