Qian Liu

Title: Qian Liu and Event-Driven Spiking Neural Networks

Introduction

Qian Liu is an inventor based in Otelfingen, Switzerland. His patent work includes an event-driven spiking convolutional neural network, a design that processes neural activity through incoming spike events.

Latest Patents

Liu is associated with 1 patent, titled “Event-Driven Spiking Convolutional Neural Network.” The invention describes a neural network with multiple layers, each containing a kernel module for storing and processing convolution-kernel values, a neuron module for managing neuron states, and a memory mapper for determining which neurons receive incoming spike events. The mapper processes those events in an event-driven manner, and the selected neurons’ states are updated using the applicable kernel values.

Career Highlights

The patent highlights Liu’s work on event-driven neural-network architecture, combining convolution operations with mechanisms for managing spike events and neuron states.

Collaborations

Liu’s listed coworkers include Ole Juri Richter and Ning Qiao.

Conclusion

Qian Liu’s patent presents an event-driven approach to spiking convolutional neural networks, with coordinated modules for processing kernel values, spike events, and neuron states.