Neuromorphic computing recently gained a lot of interest to enhance the performance and efficiency of neural network inference and training. In this presentation, I will discuss the prospects of analog signal processing in general and present examples of integrated photonic implementations. What are the advantages and challenges of photonic structures for the application in accelerators for neuromorphic computing? What type of operations are especially suited to be performed in the optical domain?
Bert Offrein received his Ph.D. degree in nonlinear integrated optics from the University of Twente (NL) in 1994. He then joined IBM Research Europe - 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 40 patents. Since 2021, Bert Offrein is professor at the University of Twente, in addition to his responsibilities at IBM.