Learning through ferroelectric domain dynamics in solid-state synapses

TLDR

Synaptic plasticity in the brain enables learning by adjusting connection strengths, and solid‑state memristors can emulate this by tuning conductance with voltage pulses following spike‑timing‑dependent plasticity (STDP). The authors seek to understand the physical mechanisms that enable STDP in nanosynapses, a prerequisite for scaling neuromorphic architectures to billions of devices. Using scanning probe imaging, electrical transport measurements, and atomistic molecular dynamics, they show that conductance changes arise from nucleation‑dominated domain reversal in ferroelectric tunnel junctions. Simulations based on this model demonstrate that arrays of ferroelectric nanosynapses can autonomously learn to recognize patterns in a predictable manner, paving the way for unsupervised learning in spiking neural networks.

Abstract

Abstract In the brain, learning is achieved through the ability of synapses to reconfigure the strength by which they connect neurons (synaptic plasticity). In promising solid-state synapses called memristors, conductance can be finely tuned by voltage pulses and set to evolve according to a biological learning rule called spike-timing-dependent plasticity (STDP). Future neuromorphic architectures will comprise billions of such nanosynapses, which require a clear understanding of the physical mechanisms responsible for plasticity. Here we report on synapses based on ferroelectric tunnel junctions and show that STDP can be harnessed from inhomogeneous polarization switching. Through combined scanning probe imaging, electrical transport and atomic-scale molecular dynamics, we demonstrate that conductance variations can be modelled by the nucleation-dominated reversal of domains. Based on this physical model, our simulations show that arrays of ferroelectric nanosynapses can autonomously learn to recognize patterns in a predictable way, opening the path towards unsupervised learning in spiking neural networks.