Zdenka Kuncic is a Professor of Physics at the University of Sydney, Australia, where she has been a faculty member for over 25 years. She earned her PhD in Theoretical Astrophysics at the University of Cambridge, UK. Her current research spans multiple areas of physics, including quantum imaging, medical nanotechnology and physical neuromorphic computing. Prof. Kuncic has authored more than 150 peer-reviewed publications and delivered numerous invited talks at leading international conferences, including the exclusive Nature Conference on Neuromorphic Computing. Her work has been recognized with several major awards, such as the Australian-American Fulbright Senior Research Scholar Award, and the Australian-Harvard Research Fellowship.

Metal-dielectric nanowire networks exhibit remarkable brain-like dynamics under electrical stimulation due to the interactions between millions of memristive junctions connected in a brain-like topology [1]. In this talk, I will review the neuromorphic properties of nanowire networks and show how their nonlinear dynamics can be exploited for reservoir computing (RC). I will also highlight how RC with physical systems like nanowire networks differs from conventional algorithmic RC, where dynamics is introduced with a mathematical activation function, rather than being intrinsic to the system, as in physical reservoirs. I will show results (e.g. Fig. 1) that demonstrate how the intrinsic dynamics of nanowire networks can be harnessed for learning and optimized for different learning tasks.

1. Z. Kuncic, T. Nakayama, Advances in Physics: X. 6(1), 1894234 (2021) DOI: 10.1080/23746149.2021.1894234
2. R. Zhu, S. Lilak, A. Loeffler, J. Lizier, A. Stieg, J. Gimzewski, Z. Kuncic, Nature Communications 14, 6697(2023) DOI: 10.1038/s41467-023-42470-5