Hardware calibrated learning to compensate heterogeneity in analog RRAM-based Spiking Neural Networks

Filippo Moro (1) , E. Esmanhotto (1) , T. Hirtzlin (1) , N. Castellani (1) , A. Trabelsi (1) , T. Dalgaty (2) , G. Molas (1) , F. Andrieu (1) , S. Brivio , S. Spiga , G. Indiveri , M. Payvand , E. Vianello (1)

1 DCOS - Département Composants Silicium
2 LIIM - Laboratoire Intelligence Intégrée Multi-capteurs

Domaines: Sciences de l'ingénieur [physics]

Filippo Moro, E. Esmanhotto, T. Hirtzlin, N. Castellani, A. Trabelsi, et al.. Hardware calibrated learning to compensate heterogeneity in analog RRAM-based Spiking Neural Networks. 2022 IEEE International Symposium on Circuits and Systems (ISCAS), May 2022, Austin, France. pp.380-383, ⟨10.1109/ISCAS48785.2022.9937820⟩. ⟨hal-04458350⟩