IBM Research funded by DARPA to mimic Human Brain on Chip – SyNAPSE Project

IBM research still heavily invests on new programs that drives innovation in the Industry. In the recent research to mimic the best parts of human brain function on a highly intelligent computer to decypher tons of data quickly, it got investors.

IBM received $16.1 million to kick start its’ part of a DARPA – Defense Advanced Research Projects Agency research program aimed at “Rapidly and efficiently put brain-like senses into actual hardware and software so that computers can process and understand data more rapidly”.

IBM has now got $21 million to work on the program known as Systems of neuromorphic adaptive plastic scalable electronics (SyNAPSE) which includes researchers from HRL Laboratories, which got $16.2 million in Oct. 2008, and others such as HP.

According to DARPA,

“The SyNAPSE program will create useful, intelligent machines. In DARPA language: the agency is looking to develop electronic neuromorphic machine technology that is scalable to biological levels. The goal is to develop systems capable of analyzing vast amounts of data from many sources in the blink of an eye, letting the military or civilian businesses make rapid decisions in time to have a significant impact on a given problem or situation. Currently, Programmable machines are limited not only by their computational capacity, but also by an architecture requiring (human-derived) algorithms to both describe and process information from their environment. In contrast, biological neural systems such as human brains, autonomously process information in complex environments by automatically learning relevant and probabilistically stable features and associations.”

As compared to biological systems for example, today’s programmable machines are less efficient by a factor of one million to one billion in complex, real-world environments. The SyNAPSE program seeks to break the programmable machine archetype and define a new path forward, DARPA stated.

DARPA goes on to state that realizing this ambitious goal will require the collaboration of numerous technical disciplines such as computational neuroscience, artificial neural networks, large-scale computation, neuromorphic VLSI, information science, cognitive science, materials science, unconventional nanometer-scale electronics, and CMOS design and fabrication.

VN:F [1.9.22_1171]
Rating: 0.0/10 (0 votes cast)
VN:F [1.9.22_1171]
Rating: 0 (from 0 votes)
Filed in: Hardware Tags: , , ,

Related Posts

Bookmark and Promote!

  • Bupress

    Boston University scientists unveil roadmap to build intelligent machines with silicon synapses
    — In a featured article to appear on the IEEE Spectrum December issue, Massimiliano Versace and Ben Chandler explain how the memristor-based approach to AI will allow to build brain-scale chips that mimics how neurons process information.—

    BOSTON, MA — November 29, 2010 – Massimiliano Versace and Ben Chandler , two scientists of the Neuromorphic Lab, Department of Cognitive and Neural Systems, Boston University, are featured on the cover page on the December issue of IEEE Spectrum , the publication of world's largest professional technology association. The feature article, appeared online on November 34, 2010, describes the ongoing effort at Boston University in building brain-scale neural models to power the next generation, low power, massively parallel chips to be realized in the DARPA SyNAPSE project in collaboration with Hewlett-Packard.

    The DARPA sponsored SyNAPSE (Systems of Neuromorphic Adaptive Plastic Scalable Electronics) project, launched in early 2009, aims to “investigate innovative approaches that enable revolutionary advances in neuromorphic electronic devices that are scalable to biological levels.” DARPA has awarded funds to three prime contractors (HP, IBM, and HRL), with HP and HRL working with Boston University researchers in the CELEST center, where the Neuromorphics Lab is housed.

    In the article, Versace and Chandler talk about recent trend in bio-inspired computing, and how these are going to shape the future of computing beyond neuroscience. In particular, the article explain how a revolutionary technology based on memristors is enabling the manufacturing of paradigm changing devices, allowing to implement low power, dense memories closer to where computation occurs, decreasing wiring length, power dissipation, and enabling to build large-scale, low power, portable devices that implement intelligent computation.

    More information on this project is available on the Neurdon blog, started by students and postdocs at Boston University, who has rapidly become a central hub in computational neuroscience and neuromorphic technology.

    VA:F [1.9.22_1171]
    Rating: 0.0/5 (0 votes cast)
    VA:F [1.9.22_1171]
    Rating: 0 (from 0 votes)
© 2018 Geeknizer. All rights reserved. XHTML / CSS Valid.
Designed by taranfx.