Collective dynamics and memory-induced long-range order in spiking oscillator arrays
ORAL
Abstract
Recent work has experimentally demonstrated a new class of spiking oscillators known as "thermal neuristors," [1] which operate through thermal interactions between adjacent vanadium dioxide resistive memories. Using numerical simulations, we demonstrate that an array of these thermal neuristors exhibits a wide range of collective behaviors. By adjusting the memory strength and external inputs, we observe a variety of phases, including strong correlations, scale-free clusters, disordered states, as well as bifurcations and chimera states. Our analysis reveals that memory plays a crucial role in shaping these collective dynamics, and we provide an analytical understanding that, despite local interactions, an adequate amount of memory results in long-range order across the array. Our findings pave the way for future experiments aimed at observing these intriguing states.
–
Publication: [1] Qiu, E., Zhang, Y. H., Di Ventra, M., & Schuller, I. K. (2023). Reconfigurable cascaded thermal neuristors for neuromorphic computing. arXiv preprint arXiv:2307.11256.
[2] Zhang, Y. H., Sipling. C. and Di Ventra, M. (2023). Collective dynamics and memory-induced long-range order in spiking oscillator arrays. In preparation.
Presenters
-
Yuan-Hang Zhang
University of California, San Diego
Authors
-
Yuan-Hang Zhang
University of California, San Diego
-
Massimiliano Di Ventra
University of California, San Diego