Session 2 – 29 April

Neuromorphic pattern recognition for the sensor edge

16:15 – 16:55 

Processing sensor data close to the sensor is challenging on account of the narrow power-area-latency envelope of typical sensing applications. In this talk, Sumeet Kumar explores how the brain processes sensory data and identifies the key to its efficiency – the notion of time. He provides a preview of how Innatera leverages fine-grained temporal processing with brain-inspired spiking neural networks using its revolutionary analog-mixed signal architecture and shares early results from its application to use cases at the sensor edge.

Sumeet Kumar is CEO of Innatera Nanosystems, the pioneering Dutch neuromorphic processor company. He holds an MSc and PhD in microelectronics from Delft University of Technology. 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, he’s credited with creating two highly successful European R&D programs developing energy-efficient compute hardware for highly automated vehicles, together with organizations including Infineon, NXP and BMW. He was also responsible for leading industry-focused research on power-efficient multiprocessors and computational neuroscience.

Sumeet Kumar

CEO at Innatera Nanosystems