Thomas Ivan Borchert

Title: Innovations of Thomas Ivan Borchert

Introduction

Thomas Ivan Borchert is a notable inventor based in Cambridge, GB. He has made significant contributions to the field of online learning algorithms and event prediction. With a total of 3 patents, his work has had a considerable impact on technology and data processing.

Latest Patents

One of his latest patents is titled "Parallelization of online learning algorithms." This patent describes methods, systems, and media for a dynamic batch strategy utilized in the parallelization of online learning algorithms. The dynamic batch strategy provides a merge function based on a threshold level difference between the original model state and an updated model state. This innovative approach allows for the reading of incoming streaming data, retrieving missing model beliefs from partner processors, and training on the batch of data until the measured difference in states exceeds a set threshold level. The merging process includes combining the merged differences with the original partial model states to obtain an updated global model state.

Another significant patent is "Event prediction in dynamic environments." This invention describes a prediction engine that uses learned information to predict events for various applications, such as internet advertising, email filtering, and fraud detection. The prediction engine considers pre-specified features associated with events, each having an associated weight and time stamp. The uncertainty about the influence of these variables on future event predictions is managed through a dynamics process that modifies the probability distribution based on the time stamp.

Career Highlights

Thomas Ivan Borchert is currently associated with Microsoft Technology Licensing, LLC, where he continues to innovate and develop new technologies. His work has been instrumental in advancing the capabilities of machine learning and predictive analytics.

Collaborations

Throughout his career, Thomas has collaborated with notable colleagues, including Ralf Herbrich and Thore Graepel. These collaborations have further enriched his contributions to the field.

Conclusion

Thomas Ivan Borchert's innovative work in online learning algorithms and event prediction showcases his expertise and commitment to advancing technology. His patents reflect a deep understanding of complex systems and their applications in real-world scenarios.