Cargando…

Impact of Asymmetric Weight Update on Neural Network Training With Tiki-Taka Algorithm

Recent progress in novel non-volatile memory-based synaptic device technologies and their feasibility for matrix-vector multiplication (MVM) has ignited active research on implementing analog neural network training accelerators with resistive crosspoint arrays. While significant performance boost a...

Descripción completa

Detalles Bibliográficos
Autores principales: Lee, Chaeun, Noh, Kyungmi, Ji, Wonjae, Gokmen, Tayfun, Kim, Seyoung
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8770851/
https://www.ncbi.nlm.nih.gov/pubmed/35069098
http://dx.doi.org/10.3389/fnins.2021.767953