The cat is out of the bag: cortical simulations with 109 neurons, 1013 synapses

By R. Ananthanarayanan, Steven K. Esser, Horst D. Simon, Dharmendra S. Modha. Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, 2009

302 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/c8d4c8870dc0adc63e196da1fd07b8aa525b4bbf

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Optoelectronic Reservoir Computing (2011, background): “For a presentation of the simulation, see [32] [40] Machine learning is the field of artificial intelligence concerned with developing algorithms whose response to the input -typically empirical data- improves with experience.”
An Electronic Synapse Device Based on Metal Oxide Resistive Switching Memory for Neuromorphic Computation (2011, background): “resources, requiring clusters of 147 456 microprocessors and 144 TB of memories, which consume a power of 1.4 MW [ 1 ], [2].”
Overview of the SpiNNaker System Architecture (2013, methodology): “An IBM project, funded under the DARPA SyNAPSE programme [12] claims the successful modeling of a neural network on the scale of a cat cortex (which is around a billion neurons with 10(13) synapses).”
A digital neurosynaptic core using embedded crossbar memory with 45pJ per spike in 45nm (2011, background): “While it has become possible to simulate large-scale brain-like networks using modern day supercomputers [1], these simulations are many orders of magnitude less efficient in power and space than biological systems.”
NeuroSim: A Circuit-Level Macro Model for Benchmarking Neuro-Inspired Architectures in Online Learning (2018, background): “However, this approach is limited due to significantly higher power consumption to mimic a biological brain with ∼1013 synapses, which can only be feasible with a very powerful supercomputer [3].”
A 45nm CMOS neuromorphic chip with a scalable architecture for learning in networks of spiking neurons (2011, methodology): “A massively scaled example of this work is described in [3], which reports a cortical simulation at the scale of a cat cortex (10(9) neurons and 10(13) synapses) using supercomputers.”
Living Together: Mind and Machine Intelligence (2017, background): “One estimate from Ananthanarayanan et al [3] suggests that it would require over an exaflop5 to perform a full simulation of the human brain.”
TrueNorth: Design and Tool Flow of a 65 mW 1 Million Neuron Programmable Neurosynaptic Chip (2015, unknown): “and “rat-scale” on Blue Gene/L [7] to “cat-scale” on Blue Gene/P [8] to “human-scale” on LLNL’s Sequoia 96-rack Blue Gene/Q [9], [10].”