Neuromorphic Hardware – A System Perspective

This seminar talk explores neuromorphic computing from a system perspective, highlighting its potential to reduce energy consumption in AI applications. Learn about key principles of neuromorphic hardware systems and their practicability as presented by Johannes Partzsch from TU Dresden. Discover the SpiNNaker2 system currently deployed at TU Dresden, one of the few large-scale neuromorphic systems in existence today. The presentation also provides an outlook on future developments in neuromorphic hardware and their connection to recent AI trends. Dr. Partzsch, group leader at the chair of Highly-Parallel VLSI Systems and Neuromicroelectronics, shares insights from his research on neuromorphic hardware development, numerical function accelerators, and edge AI hardware, including his work leading the award-winning ZEN project team and coordinating the construction of the "SpiNNcloud" neuromorphic supercomputer.