Aleksandra KOROLEVA

Abstract: This project focuses on the use of La2NiO4 devices as artificial synapses for bio-inspired computing architectures, i.e., Spiking Neural Networks (SNNs) with a bio-inspired learning rule. The learning is applied on each synapse independently of the global state of the network; therefore, the synapse must be doted of computation capabilities. The goal of the project is to demonstrate, at small scale, the feasibility of a La2NiO4-based SNN and understand the main advantages and shortcomings (from technology and application) such that concomitant optimization of device and algorithm can be performed to guarantee the achievement of a truly efficient bio-inspired electronic system.