Deep Learning Seminar / 15. Oktober 2020, 10:00 – 11:00 Uhr
The Spiking Neural Network & the Accelerators: A Polyamorous Affair
Gastreferent: Dr. Ir. Christos Strydis
[nur in Englisch verfügbar]
Spiking Neural Networks (SNNs) have been touted as wondrous new constructs for reinvigorating the fields of computational neuroscience as well as artificial intelligence and, more surreptitiously, modern computer architecture. However, their benefits are yet to be reaped while the computational challenges they introduce, as compared to traditional Artificial Neural Networks (ANNs), are significant. In this talk, I will take the audience down a ten-year journey of exploring these SNN challenges, typical community fallacies and obsessions and my solutions to the hard-posed SNN problem. With luck, newcomers to the field with learn from my mistakes and do one better.
Dr. Christos Strydis is a tenured assistant professor with the Neuroscience department of the Erasmus Medical Center, the Netherlands.
He is the founder and head of the NeuroComputing Laboratory and a senior member of the IEEE and is also a chief engineer with Neurasmus BV, the Netherlands. Dr. Strydis studied Electronics & Computer Engineering at the Technical University of Crete, Greece, and in 2003 received his bachelor's diploma (magna cum laude). In 2005, he obtained his M.Sc. degree (magna cum laude) in Computer Engineering from the Delft University of Technology, The Netherlands, with a minor in Biomedical Engineering. In 2011, he obtained his Ph.D. degree in Computer Engineering from the Delft University of Technology and funding by the ICT Delft Research Centre (DRC-ICT) and Google Inc. Christos has acted as technical-program-committee member in various international conferences. He has also peer-reviewed for as well as published manuscripts in well-known international conferences and journals, and delivered invited talks in various venues. He has been awarded a number of national- and EU-level research projects. Christos has supervised multiple BSc, MSc and PhD students, and teaches several bachelor- and master-level courses. His current research interests span the fields of brain simulations, high-performance computing, low-power embedded (in particular, implantable) systems and functional-ultrasound brain imaging.