A TeraMAC Neuromorphic Photonic Processor

We show that an integrated laser neuron can exhibit extraordinary low latency (<1ns) and speed (∼1 × 1012 MACs/s per device) compared to state-of-the-art processors in digital electronics. We experimentally demonstrate positive (excitatory) and negative (inhibitory) inputs with 8× wavelength channels, and efficiency (<1pJ/MAC) during closed-loop operation.

[1]  Dirk Englund,et al.  Deep learning with coherent nanophotonic circuits , 2017, 2017 Fifth Berkeley Symposium on Energy Efficient Electronic Systems & Steep Transistors Workshop (E3S).

[2]  David A. Patterson,et al.  In-datacenter performance analysis of a tensor processing unit , 2017, 2017 ACM/IEEE 44th Annual International Symposium on Computer Architecture (ISCA).

[3]  Paul R Prucnal,et al.  Two-pole microring weight banks. , 2018, Optics letters.

[4]  P. Prucnal,et al.  NEUROMORPHIC PHOTONICS , 2017 .

[5]  P. R. Prucnal,et al.  A Leaky Integrate-and-Fire Laser Neuron for Ultrafast Cognitive Computing , 2013, IEEE Journal of Selected Topics in Quantum Electronics.

[6]  Paul R. Prucnal,et al.  Spike processing with a graphene excitable laser , 2016, Scientific Reports.

[7]  David A. B. Miller.  Attojoule Optoelectronics for Low-Energy Information Processing and Communications , 2017, Journal of Lightwave Technology.

[8]  Paul R. Prucnal,et al.  Broadcast and Weight: An Integrated Network For Scalable Photonic Spike Processing , 2014, Journal of Lightwave Technology.