Bio-inspired AI from Edge to Cloud: HW-SW Codesign
![Printer-friendly version Printer-friendly version](/sites/all/modules/print/icons/print_icon.gif)
![Send by email Send by email](/sites/all/modules/print/icons/mail_icon.gif)
![PDF version PDF version](/sites/all/modules/print/icons/pdf_icon.gif)
Christian Mayr (TU Dresden, D)
Abstract
AI is having an increasingly large impact on our daily lives. However, current AI hardware and algorithms are still only partially inspired by the major blueprint for AI, i.e. the human brain. In particular, even the best AI hardware is still far away from the 20W power consumption, the low latency and the unprecedented large scale, high-throughput processing offered by the human brain.
In this talk, I will describe our bio-inspired AI hardware, from our award-winning edge nodes up to our cloud-scale SpiNNaker2 system, which achieves a unique fusion of GPU, CPU and neuromorphic components. It takes inspiration from biology not just at the single-neuron level like current neuromorphic chips, but throughout all architectural levels.
SpiNNaker2 represents the largest cloud system for real-time AI worldwide. Combining it with our domain-specific edge nodes, very powerful and efficient edge-to-cloud system can be developed. Applications for these edge-to-cloud system range from smart city, industry 4.0, robotics, in-network computing for the tactile internet all the way to terrestrial and orbital situational awareness systems.
Curriculum Vitae
Christian Mayr holds the chair of Highly Parallel VLSI Systems and Neuro-Microelectronics at Technische Universität Dresden. He received the M.Sc. in Electrical Engineering in 2003, PhD in 2008, and Habilitation (university teaching qualification) in 2012, all three from Technische Universität Dresden, Germany. From 2003 to 2013, he was Chair of Highly Parallel VLSI Systems and Neuro-Microelectronics at Technische Universität Dresden, with a secondment to Infineon AG Munich (2004–2006). From 2013 to 2015, he was working as researcher at Institute of Neuroinformatics, University of Zurich and ETH Zurich, Switzerland. His research interests include bio-inspired circuits, biologically inspired artificial intelligence, brain–machine interfaces, A-to-D converters, pixel sensors, and general mixed-signal VLSI design. He is author or coauthor of over 200 publications and holds three patents. His work has received the Heinrich Barkhausen Price (2008), the Meyer Struckmann Science Price (2013), the 1Mio BMBF award for an edge-AI system, as well as various other recognitions.