PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
August 7, 2014Science4,181 citations

A million spiking-neuron integrated circuit with a scalable communication network and interface

View Full Paper
PMPaul MerollaJAJohn V. ArthurRARodrigo Alvarez-Icaza

Key Points

  • To develop a scalable, energy-efficient, and flexible non-von Neumann chip architecture inspired by biological neural networks.
  • Fabricated a 5.4-billion-transistor silicon integrated circuit comprising 4,096 neurosynaptic cores.
  • Integrated 1 million programmable spiking neurons and 256 million configurable synapses via an intrachip network and a 2D scalable interchip interface.
  • Demonstrated seamless 2D interchip tiling to scale the architecture into arbitrarily large cortex-like computing sheets.
  • Achieved real-time multi-object detection and classification on 400x240-pixel video at 30 frames per second with a power consumption of 63 milliwatts.

Abstract

Inspired by the brain's structure, we have developed an efficient, scalable, and flexible non-von Neumann architecture that leverages contemporary silicon technology. To demonstrate, we built a 5.4-billion-transistor chip with 4096 neurosynaptic cores interconnected via an intrachip network that integrates 1 million programmable spiking neurons and 256 million configurable synapses. Chips can be tiled in two dimensions via an interchip communication interface, seamlessly scaling the architecture to a cortexlike sheet of arbitrary size. The architecture is well suited to many applications that use complex neural networks in real time, for example, multiobject detection and classification. With 400-pixel-by-240-pixel video input at 30 frames per second, the chip consumes 63 milliwatts.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Merolla et al. (2014) studied this question.

synapsesocial.com/papers/69d68bdb96200ba434db827chttps://doi.org/10.1126/science.1254642
Ask AI
Helpful
Bookmark
Share
View Full Paper