Jaspreet Sahota

Title: Innovations of Jaspreet Sahota

Introduction

Jaspreet Sahota is an accomplished inventor based in Toronto, Canada. He has made significant contributions to the field of machine learning and predictive analytics. With a total of 3 patents to his name, his work focuses on enhancing the accuracy and efficiency of machine learning systems.

Latest Patents

One of his latest patents is the "K-LSTM architecture for purchase prediction." This innovative system receives transaction data over time and creates structured data based on the received transaction data. It identifies purchase transactions associated with specific categories and labels them accordingly. A recurrent neural network, particularly a k-LSTM architecture, is trained to build a model that predicts the likelihood of a purchase transaction.

Another notable patent is the "System and method for time-dependent machine learning architecture." This technical solution addresses the decomposition of time as an input for machine learning. It includes various mechanisms and data structures aimed at improving machine learning outcomes, such as accuracy and convergence speed. The invention utilizes a vector representation of continuous time, which contains learnable parameters for frequency and phase-shift, to enhance the tracking of periodic behavior in features.

Career Highlights

Jaspreet currently works at the Royal Bank of Canada, where he applies his expertise in machine learning to develop innovative solutions. His work has been instrumental in advancing the bank's capabilities in predictive analytics and data-driven decision-making.

Collaborations

He collaborates with talented individuals such as Janahan Mathuran Ramanan and Yuanqiao Wu, contributing to a dynamic and innovative work environment.

Conclusion

Jaspreet Sahota's contributions to machine learning and predictive analytics through his patents demonstrate his commitment to innovation. His work continues to influence the field and drive advancements in technology.