Pulse nav.journalClub Debates activos Tendencias Explorar Investigadores

Download the App

Join discussions, follow papers, and never miss your next session.

Download on theApp Store

© Synapse Social LLC, 2026

Política de privacidad

Inicio Explorar nav.journalClub Tendencias

⌘+K

A domain-specific architecture for deep neural networks | Synapse

August 22, 2018Open Access

A domain-specific architecture for deep neural networks

Puntos clave

Los puntos clave no están disponibles para este artículo en este momento.

Resumen

Tensor processing units improve performance per watt of neural networks in Google datacenters by roughly 50x.

Me gusta

Guardar

Compartir

Ver artículo completo

Me gusta

Guardar

Compartir

Ver artículo completo

Cite This Study

Jouppi et al. (Wed,) studied this question.

synapsesocial.com/papers/6a1be7f726cb5670aa9d1892 https://doi.org/https://doi.org/10.1145/3154484

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

1Technical Perspective: If I could only design one circuit …2016 · 5 citations
2Identifying Shades of Green: The SPECpower Benchmarks2009 · 108 citations
3No exponential Is forever : But "Forever" can be delayed2003 · 179 citations
4The Case for Energy-Proportional Computing2007 · 2,479 citations
5The case for the reduced instruction set computer1980 · 211 citations