Ergodicity, lack thereof, and the performance of reservoir computing with memristive networks

Abstract

Networks composed of nanoscale memristive components, such as nanowire and nanoparticle networks, have recently received considerable attention because of their potential use as neuromorphic devices. In this study, we explore the connection between ergodicity in memristive and nanowire networks, showing that the performance of reservoir devices improves when these networks are tuned to operate at the edge between two global stability points. The lack of ergodicity is associated with the emergence of memory in the system. We measure the level of ergodicity using the Thirumalai-Mountain metric, and we show that in the absence of ergodicity, two memristive systems show improved performance when utilized as reservoir computers (RC). In particular, we highlight that it is also important to let the system synchronize to the input signal in order for the performance of the RC to exhibit improvements over the baseline.

10 figures

References in corpus (11)

• The elusive memristor: properties of basic electrical circuits
• Large-scale photonic Ising machine by spatial light modulation
• Intrinsic optimization using stochastic nanomagnets
• Ergodicity in Natural Earthquake Fault Networks
• Ergodicity Breaking in a Deterministic Dynamical System
• Direct observation of a dynamical glass transition in a nanomagnetic artificial Hopfield network
• The complex dynamics of memristive circuits: analytical results and universal slow relaxation
• Complex dynamics and scale invariance of one-dimensional memristive networks
• Mean field theory of self-organizing memristive connectomes
• Effective ergodicity in single-spin-flip dynamics
• Projective Embedding of Dynamical Systems: uniform mean field equations

Cited by in corpus (4)

• Learning Chaotic Dynamics with Neuromorphic Network Dynamics
• The dipolar Aleppo lattice: Ground state ordering and ergodic dynamics in the absence of vertex frustration
• Ergodicity of FIRE: star formation variations within and between simulated galaxies
• Training-free AI for Earth Observation Change Detection using Physics Aware Neuromorphic Networks