Tree-based machine learning performed in-memory with memristive analog CAM

Tree-based machine learning algorithms are known to be explainable and effective even trained on limited datasets, however difficult to optimize on conventional digital hardware. The authors apply analog content addressable memory to accelerate tree-based model inference for improved performance.

Guardado en:
Detalles Bibliográficos
Autores principales: Giacomo Pedretti, Catherine E. Graves, Sergey Serebryakov, Ruibin Mao, Xia Sheng, Martin Foltin, Can Li, John Paul Strachan
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
Q
Acceso en línea:https://doaj.org/article/2918fbbcb63643d0ba6ac994f9f06593
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Descripción
Sumario:Tree-based machine learning algorithms are known to be explainable and effective even trained on limited datasets, however difficult to optimize on conventional digital hardware. The authors apply analog content addressable memory to accelerate tree-based model inference for improved performance.