Always-on sub-milliwatt AI for the sensor edge | Kisaco Research

Spiking neural networks are central to the brain’s processing of sensory data, and are key to its ability to recognize spatial and time-series patterns quickly and efficiently.

Innatera’s Spiking Neural Processor (SNP) enables always-on AI applications based on programmable, energy-efficient acceleration of spiking neural networks. It achieves unprecedented AI-based functions at the sensor-edge within a sub-milliwatt power and sub-millisecond latency envelope, using spiking models 100x smaller than conventional neural networks. In this talk, CEO Dr. Sumeet Kumar explores the capabilities of the SNP, and how these link with the critical requirements of sensing use-cases in the consumer and industrial domains. Further, Dr. Kumar provides insight into Innatera’s Talamo software development kit which brings the power and familiarity of PyTorch to spiking neural networks.

Speaker(s): 

Author:

Sumeet Kumar

CEO
Innatera Nanosystems

Dr. Sumeet Kumar is CEO of Innatera Nanosystems, the pioneering Dutch neuromorphic processor company. Dr. Kumar holds an MSc and PhD in Microelectronics from the Delft University of Technology, The Netherlands. He was previously with Intel, where he worked with the Imaging and Camera Technologies Group developing domain-specific tools for the development of complex media processor architectures. At Delft, Dr. Kumar is credited with creating two highly-successful European R&D programmes developing energy-efficient compute hardware for highly automated vehicles, together with organizations including Infineon, NXP, and BMW, among others. He was also responsible for leading industry-focused research on power-efficient multiprocessors and computational neuroscience.

Sumeet Kumar

CEO
Innatera Nanosystems

Dr. Sumeet Kumar is CEO of Innatera Nanosystems, the pioneering Dutch neuromorphic processor company. Dr. Kumar holds an MSc and PhD in Microelectronics from the Delft University of Technology, The Netherlands. He was previously with Intel, where he worked with the Imaging and Camera Technologies Group developing domain-specific tools for the development of complex media processor architectures. At Delft, Dr. Kumar is credited with creating two highly-successful European R&D programmes developing energy-efficient compute hardware for highly automated vehicles, together with organizations including Infineon, NXP, and BMW, among others. He was also responsible for leading industry-focused research on power-efficient multiprocessors and computational neuroscience.