Stochastic resonance effect in binary STDP performed by RRAM devices

The beneficial role of noise in the binary spike time dependent plasticity (STDP) learning rule, when implemented with memristors, is experimentally analyzed. The two memristor conductance states, which emulate the neuron synapse in neuromorphic architectures, can be better distinguished if a gaussi...

Descripción completa

Detalles Bibliográficos
Autores: Salvador Aguilera, Emili|||0000-0002-1613-6784, Rodríguez Martínez, Rosana|||0000-0002-4565-6703, Martin Martinez, Javier|||0000-0001-5938-5898, Crespo Yepes, Albert|||0000-0003-4618-651X, Miranda, E.|||0000-0003-0470-5318, Nafria, Montserrat|||0000-0002-9549-2890, Rubio, Antonio|||0000-0003-1625-1472, Ntinas, Vasileios|||0000-0002-2367-5567, Sirakoulis, Georgios Ch|||0000-0001-8240-484X
Tipo de recurso: capítulo de libro
Fecha de publicación:2022
País:España
Institución:Universitat Autònoma de Barcelona
Repositorio:Dipòsit Digital de Documents de la UAB
Idioma:inglés
OAI Identifier:oai:ddd.uab.cat:291869
Acceso en línea:https://ddd.uab.cat/record/291869
https://dx.doi.org/urn:doi:10.1109/NANO54668.2022.9928738
Access Level:acceso abierto
Palabra clave:Memristor
Neuromorphic systems
Resistive switching
RRAM
STDP
Stochastic resonance
Descripción
Sumario:The beneficial role of noise in the binary spike time dependent plasticity (STDP) learning rule, when implemented with memristors, is experimentally analyzed. The two memristor conductance states, which emulate the neuron synapse in neuromorphic architectures, can be better distinguished if a gaussian noise is added to the bias. The addition of noise allows to reach memristor conductances which are proportional to the overlap between pre- and post-synaptic pulses.

Similares en LA Referencia