This inventor holds 1 USPTO granted patent. Top assignee: Preferred Networks, Inc.. Active years: 2026.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2026
Title: Yasuhiro Fujita and Distributed Reinforcement Learning
Introduction
Yasuhiro Fujita is an inventor associated with Tokyo, Japan, and Preferred Networks, Inc. His patent work concerns distributed reinforcement learning systems and methods.
Latest Patents
Fujita is credited with 1 patent: “Distributed Reinforcement Learning System and Distributed Reinforcement Learning Method.” The invention describes a system in which actor devices collect experience data based on actions determined by a model being trained. Multiple replay buffers store that data across distributed nodes, while learner devices use the stored experience to train the model through reinforcement learning.
Career Highlights
Fujita’s listed work at Preferred Networks, Inc. is connected to the development of distributed approaches to reinforcement learning. The patent outlines how actors, replay buffers, and learner devices can work together across multiple nodes.
Collaborations
Fujita’s listed coworker is Kota Uenishi. Their association is reflected in the patent information provided.
Conclusion
Yasuhiro Fujita’s patent documents a distributed reinforcement learning approach that organizes experience collection, storage, and model training across multiple devices and nodes.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile