Wang Fumin

Title: Wang Fumin: Innovator in Neural Network Systems

Introduction

Wang Fumin is a prominent inventor based in London, GB. He has made significant contributions to the field of artificial intelligence, particularly in the development of neural network systems. With a total of 2 patents, his work focuses on enhancing the capabilities of reinforcement learning systems.

Latest Patents

Wang Fumin's latest patents include a groundbreaking system for training action selection neural networks using apprenticeship. This off-policy reinforcement learning actor-critic neural network system is designed to select actions from a continuous action space for an agent interacting with an environment to perform specific tasks. The system utilizes an actor neural network that learns a policy function, mapping environment state data to action data. Additionally, a critic neural network is employed to learn an action-value (Q) function. The system also features a replay buffer that stores tuples of state data, action data, reward data, and new state data, including demonstration transition data from task demonstrations within the environment. This innovative approach allows for off-policy training of both the actor and critic neural networks using stored tuples from the replay buffer.

Career Highlights

Wang Fumin is currently associated with DeepMind Technologies Limited, a leading company in artificial intelligence research. His work at DeepMind has positioned him as a key player in advancing neural network technologies.

Collaborations

Wang has collaborated with notable colleagues such as Olivier Claude Pietquin and Martin Riedmiller, contributing to the collective expertise in the field of artificial intelligence.

Conclusion

Wang Fumin's contributions to neural network systems exemplify the innovative spirit of modern inventors. His work continues to influence the development of advanced AI technologies, paving the way for future advancements in the field.