The challenge of developing an efficient artificial neuron is impeded by the use of external CMOS circuits to perform leaking and lateral inhibition. The proposed leaky integrate-and-fire neuron based on the three terminal magnetic tunnel junction (3T-MTJ) performs integration by pushing its domain wall (DW) with spin-transfer or spin-orbit torque. The leaking capability is achieved by pushing the neurons’ DWs in the direction opposite of integration using a stray field from a hard ferromagnet or a non-uniform energy landscape resulting from shape or anisotropy variation. Firing is performed by the MTJ stack. Finally, analog lateral inhibition is achieved by dipolar field repulsive coupling from each neuron. An integrating neuron thus pushes slower neighboring neurons’ DWs in the direction opposite of integration. Applying this lateral inhibition to a ten-neuron output layer within a neuromorphic crossbar structure enables the identification of handwritten digits with 94% accuracy.
Magnetic domain wall neuron with intrinsic leaking and lateral inhibition capability / Brigner, Wesley H.; Hassan, Naimul; Hu, Xuan; Jiang-Wei, Lucian; Akinola, Otitoaleke G.; Garcia-Sanchez, Felipe; Pasquale, Massimo; Bennett, Christopher H.; Incorvia, Jean Anne C.; Friedman, Joseph S.. - Proc. SPIE 11090, Spintronics XII, 1109001 (19 September 2019); doi: 10.1117/12.2551338:(2019), p. 128. (Intervento presentato al convegno SPIE Nanoscience + Engineering, 2019, San Diego tenutosi a California, United States) [10.1117/12.2528218].
Magnetic domain wall neuron with intrinsic leaking and lateral inhibition capability
Garcia-Sanchez, Felipe;Pasquale, MassimoFunding Acquisition
;
2019
Abstract
The challenge of developing an efficient artificial neuron is impeded by the use of external CMOS circuits to perform leaking and lateral inhibition. The proposed leaky integrate-and-fire neuron based on the three terminal magnetic tunnel junction (3T-MTJ) performs integration by pushing its domain wall (DW) with spin-transfer or spin-orbit torque. The leaking capability is achieved by pushing the neurons’ DWs in the direction opposite of integration using a stray field from a hard ferromagnet or a non-uniform energy landscape resulting from shape or anisotropy variation. Firing is performed by the MTJ stack. Finally, analog lateral inhibition is achieved by dipolar field repulsive coupling from each neuron. An integrating neuron thus pushes slower neighboring neurons’ DWs in the direction opposite of integration. Applying this lateral inhibition to a ten-neuron output layer within a neuromorphic crossbar structure enables the identification of handwritten digits with 94% accuracy.File | Dimensione | Formato | |
---|---|---|---|
spie19.pdf
solo utenti autorizzati
Tipologia:
submitted version (author’s pre-print)
Licenza:
Non Pubblico - Accesso privato/ristretto
Dimensione
271.08 kB
Formato
Adobe PDF
|
271.08 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.