John R Thorp

Title: John R. Thorp: Innovator in Data Processing Technologies

Introduction

John R. Thorp is a notable inventor based in Columbia, MD (US), recognized for his contributions to data processing technologies. With a total of 2 patents, Thorp has made significant strides in the field of computer systems and methods for partitioning data for parallel processing.

Latest Patents

Thorp's latest patents include a groundbreaking invention titled "Computer system and computerized method for partitioning data for parallel processing." This innovative system is designed to split a data space to efficiently partition data between processors or processes. The data space can be divided into sub-regions that do not necessarily align with the axes defined by the data space's parameters, utilizing a decision tree. Each non-terminal node of the decision tree can incorporate neural networks that are trained on the data, facilitating the partitioning of training data. The system allows for the training of non-terminal nodes' neural networks on one processor, while the training of leaf nodes' networks can occur on separate processors. This configuration ensures that each leaf node receives an approximately equal number of training records, optimizing the training process.

Career Highlights

Thorp is currently employed at Torrent Systems, Inc., where he continues to develop innovative solutions in data processing. His work has significantly impacted the efficiency of data classification and parameter estimation in large databases.

Collaborations

Throughout his career, Thorp has collaborated with talented individuals such as Anthony Passera and Michael James Beckerle, contributing to the advancement of technology in his field.

Conclusion

John R. Thorp's innovative work in data processing technologies exemplifies the impact of inventive thinking on modern computing. His contributions continue to shape the landscape of parallel processing and data management.