New Delhi, India

Yogish Sabharwal

This inventor holds 26 USPTO granted patents and 8 published patent applications, primarily in Machine Learning. Top assignees: International Business Machines Corporation, Universiti Brunei Darussalam. Active years: 2010-2026.

USPTO Granted Patents = 26 

% Patents Active = 50.0

Average Co-Inventor Count = 4.5

ph-index = 5

Forward Citations = 105(Granted Patents)


Location History:

  • New Delhi, IN (2010 - 2020)
  • Bengaluru, IN (2020)
  • Haryana, IN (2013 - 2021)
  • Gurgaon, IN (2015 - 2023)

Company Filing History:


Years Active: 2010-2026

Loading Chart...
Areas of Expertise:
Neural Networks
Machine Learning
Resource Allocation
Cloud Computing
High Performance Computing
Dynamic Parameterization
Image Reconstruction
Automated Scheduling
Distributed Processing
Emissions Inventory
Power Management
Optimization
26 patents (USPTO):Explore Patents

Title: Innovations and Contributions of Yogish Sabharwal

Introduction

Yogish Sabharwal is a prominent inventor based in New Delhi, India, known for his significant contributions to the field of machine learning and artificial intelligence. With a remarkable portfolio of 22 patents, Yogish has been at the forefront of innovations that enhance the capabilities of automated systems and machine learning models.

Latest Patents

Among his latest patents, Yogish has developed a groundbreaking system called "Multi-objective automated machine learning." This invention focuses on identifying multiple machine learning pipelines as Pareto-optimal solutions to optimize various objectives. The method involves receiving input data related to specific subjects of interest, defining a plurality of objectives for optimization, and utilizing machine learning models to process the data. This innovative approach enables the aggregation of objectives into single aggregated ones, ultimately leading to the selection of the most efficient ML pipeline.

Another notable patent is his work on "Accelerating inference of transformer-based models." This invention presents methods, systems, and computer program products aimed at enhancing the performance of transformer models in natural language processing tasks. The method focuses on compressing machine learning models based on specific tasks, which involves identifying and removing redundant activations, thus generating a more efficient version of the model for user applications.

Career Highlights

Yogish has an impressive career, having worked with prestigious organizations such as IBM and Universiti Brunei Darussalam. His roles at these institutions have enabled him to collaborate with leading experts in the field, thereby fostering an environment of innovation and research excellence. His contributions to these organizations have significantly impacted the landscape of machine learning technologies.

Collaborations

Throughout his career, Yogish has collaborated with notable professionals, including Thomas George and Vaibhav Saxena. These partnerships have allowed for a cross-pollination of ideas and have contributed to the successful innovative projects he has been part of.

Conclusion

Yogish Sabharwal continues to be a vital player in the field of technology, with his extensive research and innovative patents paving the way for advancements in machine learning. His dedication to improving automated systems not only showcases his expertise as an inventor but also reflects the potential of innovations that redefine the boundaries of technology.

Profile summary based on public USPTO records.
Please report any incorrect information to [email protected]
Loading…