Short-term memory of TiO₂-based electrochemical capacitors: empirical analysis with adoption of a sliding threshold

Abstract

Chemical synapses are important components of the large-scaled neural network in the hippocampus of the mammalian brain, and a change in their weight is thought to be in charge of learning and memory. Thus, the realization of artificial chemical synapses is of crucial importance in achieving artificial neural networks emulating the brain's functionalities to some extent. This kind of research is often referred to as neuromorphic engineering. In this study, we report short-term memory behaviours of electrochemical capacitors (ECs) utilizing TiO2 mixed ionic-electronic conductor and various reactive electrode materials e.g. Ti, Ni, and Cr. By experiments, it turned out that the potentiation behaviours did not represent unlimited growth of synaptic weight. Instead, the behaviours exhibited limited synaptic weight growth that can be understood by means of an empirical equation similar to the Bienenstock-Cooper-Munro rule, employing a sliding threshold. The observed potentiation behaviours were analysed using the empirical equation and the differences between the different ECs were parameterized.

Memristive Hebbian Plasticity Model: Device Requirements for the Emulation of Hebbian Plasticity Based on Memristive Devices

“…Further, the BCM model predicts different regimes of LTP and LTD depending on the state of the post-synaptic neuron, while a sliding threshold separates this into two regimes [32]. Recently, Lim et al [33] showed evidence that BCM-like plasticity can be emulated with memristive devices and derived an empirical equation in the framework of a firing-rate-model by introducing a time-depending threshold function. Here, we would like to derive an empirical plasticity model, which itself avoids an unlimited weight growth and where the sliding threshold depends on the current synaptic weight and therewith on the state variable .…”

Anatomy of Ag/Hafnia‐Based Selectors with 1010 Nonlinearity

“…Memristors are promising candidates for both neuromorphic computing and next generation nonvolatile memory applications [1][2][3][4][5][6][7] because of their scalability [3,8] , 3D stacking potential [9][10][11] , and a close resemblance to the operating characteristics of synapses [1,4,7,12,13] . These applications typically require a large crossbar array of memristors, in which sneak path currents from neighboring cells during a write or read of a target cell can severely impede the proper operation of the array.…”

Highly Compact Artificial Memristive Neuron with Low Energy Consumption

“…Recently, a drastic research paradigm shift toward memristive devices may enable a new approach of full memristive neural networks (FMNN) for achieving bioplausible neural networks with great scaling potential . Significant progress has been made on memristive synapses, which can simulate rich synaptic functionalities on a single nanodevice . Especially, by utilizing the dynamic memory switching effect, memristors has been proposed to simulate the synaptic learning rules, leading to a more biological artificial synapse.…”