NeuronFlow: a neuromorphic processor architecture for Live AI applications
Orlando Moreira1, Amirreza Yousefzadeh1, Fabian Chersi2, Gokturk Cinserin1, Rik-Jan Zwartenkot1, Ajay Kapoor1, Peng Qiao1, Peter Kievits1, Mina Khoei2, Louis Rouillard2, Aimee Ferouge1, Jonathan Tapson3 and Ashoka Visweswara1
1GrAI Matter Labs (GML), Eindhoven, Netherlands
2GrAI Matter Labs (GML), Paris, France
3GrAI Matter Labs (GML), California, USA
ABSTRACT
Neuronflow is a neuromorphic, many core, data flow architecture that exploits brain-inspired concepts to deliver a scalable event-based processing engine for neuron networks in Live AI applications. Its design is inspired by brain biology, but not necessarily biologically plausible. The main design goal is the exploitation of sparsity to dramatically reduce latency and power consumption as required by sensor processing at the Edge.
