A bakery assistant who takes the bread from the shelf just to give it to his boss who then hands it over to the customer? Rather unlikely. Instead, both work at the same time to sell the baked goods. Similarly, computer programs are more efficient if they process data in parallel rather than to calculate them one after the other. However, most programs that are applied still work in a serial manner.
The neuromorphic chip containing silicon neurons which the researchers used for their data-classifying network.
Kirchhoff Institute for Physics, Heidelberg University
Scientists from the Freie Universität Berlin, the Bernstein Center Berlin and Heidelberg University have now refined a new technology that is based on parallel data processing. In the so-called neuromophic computing, neurons made of silicon take over the computational work on special computer chips. The neurons are linked together in a similar fashion to the nerve cells in our brain. If the assembly is fed with data, all silicon neurons work in parallel to solve the problem. The precise nature of their connections determines how the network processes the data. Once properly linked, the neuromorphic network operates almost by itself. The researchers have now designed a network–a neuromorphic “program”–for this chip that solves a fundamental computing problem: It can classify data with different features. It is able to recognize handwritten numbers, or may distinguish certain plant species based on flowering characteristics.
"The design of the network architecture has been inspired by the odor-processing nervous system of insects," explains Michael Schmuker, lead author of the study. "This system is optimized by nature for a highly parallel processing of the complex chemical world." Together with work group leader Martin Nawrot and Thomas Pfeil, Schmuker provided the proof of principle that a neuromorphic chip can solve such a complex task. For their study, the researchers used a chip with silicon neurons, which was developed at the Kirchhoff Institute for Physics of Heidelberg University.
Computer programs that can classify data are employed in various technical devices, such as smart phones. The neuromorphic network chip could also be applied in super-computers that are built on the model of the human brain to solve very complex tasks. Using their prototype, the Berlin scientists are now able to explore how networks must be designed to meet the specific requirements of these brain-like computer. A major challenge will be that not even two neurons are identical – neither in silicon nor in the brain.
The Bernstein Center Berlin is part of the National Bernstein Network Computational Neuroscience in Germany. With this funding initiative, the German Federal Ministry of Education and Research (BMBF) has supported the new discipline of Computational Neuroscience since 2004 with over 170 million Euros. The network is named after the German physiologist Julius Bernstein (1835-1917).Contact:
Weitere Informationen:http://biomachinelearning.net personal website Michael Schmuker
http://www.nncn.de National Bernstein Network Computational Neuroscience
Mareike Kardinal | idw
Superfast fluorescence sets new speed record
27.07.2015 | Duke University
Two crystals are better than one
22.07.2015 | The Agency for Science, Technology and Research (A*STAR)
Researchers have developed an ultrafast light-emitting device that can flip on and off 90 billion times a second and could form the basis of optical computing.
Joint BioEnergy Institute study identifies bacterial protein that is key to protecting rice against bacterial blight
A bacterial signal that when recognized by rice plants enables the plants to resist a devastating blight disease has been identified by a multi-national team...
Researchers in the Cockrell School of Engineering at The University of Texas at Austin are one step closer to delivering smart windows with a new level of energy efficiency, engineering materials that allow windows to reveal light without transferring heat and, conversely, to block light while allowing heat transmission, as described in two new research papers.
By allowing indoor occupants to more precisely control the energy and sunlight passing through a window, the new materials could significantly reduce costs for...
Argonne scientists used Mira to identify and improve a new mechanism for eliminating friction, which fed into the development of a hybrid material that exhibited superlubricity at the macroscale for the first time. Argonne Leadership Computing Facility (ALCF) researchers helped enable the groundbreaking simulations by overcoming a performance bottleneck that doubled the speed of the team's code.
While reviewing the simulation results of a promising new lubricant material, Argonne researcher Sanket Deshmukh stumbled upon a phenomenon that had never been...
A NASA camera on the Deep Space Climate Observatory (DSCOVR) satellite has returned its first view of the entire sunlit side of Earth from one million miles away.
The color images of Earth from NASA's Earth Polychromatic Imaging Camera (EPIC) are generated by combining three separate images to create a...
23.07.2015 | Event News
10.07.2015 | Event News
25.06.2015 | Event News
28.07.2015 | Life Sciences
28.07.2015 | Materials Sciences
28.07.2015 | Ecology, The Environment and Conservation