The performance increase of established Von Neumann computing systems has been defined by the scaling of silicon chip technology according to Moore’s law as well as advances in assembly and system integration concepts. While these scaling paths run into technological and commercial challenges, these types of systems are not well suited to process the enormous amounts of unstructured data generated nowadays. Analog synaptic signal processing holds the promise for massive performance and power-efficiency enhancements in neuromorphic computing. Memristive and photonic concepts and technologies will be presented.
Bert Offrein received his Ph.D. degree in nonlinear integrated optics from the University of Twente (NL) in 1994. He then joined IBM Research - Zurich and contributed to establishing and commercializing adaptive integrated optical technology for DWDM networks. From 2004 to 2016, Bert Offrein was managing the photonics group, addressing optical interconnects for computing systems. Since 2016, he is leading the neuromorphic devices and systems group, focussing on novel hardware for neural networks. Bert Offrein is a principal research staff member at IBM Research and the co-author of over 150 publications and the co-inventor of more than 35 patents.