Peter Morris

Title: Innovations of Peter Morris in Hybrid Scheduling for Deep Learning

Introduction

Peter Morris is an accomplished inventor based in Suwon-si, South Korea. He has made significant contributions to the field of computing, particularly in the area of deep learning workloads. His innovative approach to scheduling methods has garnered attention in the tech industry.

Latest Patents

Peter Morris holds a patent for a "Hybrid scheduling method for deep learning workloads, and computing apparatus with hybrid scheduling." This patent outlines a sophisticated scheduling method performed by a computing apparatus. The method includes generating an input vector that encompasses resource status and metadata for tasks intended for parallel execution. It involves determining actions for the input vector through an actor network, performing first and second resource scheduling based on the determined actions and input vector, and evaluating the performance of both scheduling results using a critic network. Ultimately, the method selects the most effective scheduling result for each task and allocates resources accordingly. This innovative approach enhances the efficiency of deep learning processes.

Career Highlights

Peter Morris is currently employed at Samsung Electronics Co., Ltd., where he continues to push the boundaries of technology. His work at Samsung has allowed him to collaborate with other talented professionals in the field, contributing to the company's reputation for innovation.

Collaborations

Some of Peter's notable coworkers include Jinseong Kim and Hyesun Hong. Their collaborative efforts have played a crucial role in advancing the projects they work on together.

Conclusion

Peter Morris is a notable inventor whose work in hybrid scheduling methods for deep learning workloads exemplifies innovation in computing technology. His contributions are paving the way for more efficient resource management in complex computational tasks.