Learning to Approximate Functions Using Nb-Doped SrTiO3 Memristors

Frontiers in Neuroscience

Thomas F. Tiotto¹  Jelmer P. Borst¹  Niels A. Taatgen¹  Anouk S. Goossens³  Tamalika Banerjee³ 

[1] Artificial Intelligence, Bernoulli Institute, University of Groningen, Groningen, Netherlands;Groningen Cognitive Systems and Materials Center, University of Groningen, Groningen, Netherlands;Zernike Institute for Advanced Materials, University of Groningen, Groningen, Netherlands;

关键词: neuromorphic computing;    supervised learning;    interface memristor;    Nb-doped SrTiO3;    neural networks;    spiking neural network;   

DOI  :  10.3389/fnins.2020.627276

来源: DOAJ

【 摘 要 】

Memristors have attracted interest as neuromorphic computation elements because they show promise in enabling efficient hardware implementations of artificial neurons and synapses. We performed measurements on interface-type memristors to validate their use in neuromorphic hardware. Specifically, we utilized Nb-doped SrTiO3 memristors as synapses in a simulated neural network by arranging them into differential synaptic pairs, with the weight of the connection given by the difference in normalized conductance values between the two paired memristors. This network learned to represent functions through a training process based on a novel supervised learning algorithm, during which discrete voltage pulses were applied to one of the two memristors in each pair. To simulate the fact that both the initial state of the physical memristive devices and the impact of each voltage pulse are unknown we injected noise into the simulation. Nevertheless, discrete updates based on local knowledge were shown to result in robust learning performance. Using this class of memristive devices as the synaptic weight element in a spiking neural network yields, to our knowledge, one of the first models of this kind, capable of learning to be a universal function approximator, and strongly suggests the suitability of these memristors for usage in future computing platforms.

【 授权许可】

Unknown