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
NASA's ICESat-2 equipped with unique 3-D manufactured part
03.02.2016 | NASA/Goddard Space Flight Center
Energy-saving minicomputers for the ‘Internet of Things’
29.01.2016 | Helmholtz-Zentrum Dresden-Rossendorf
Automobiles increase the mobility of their users. However, their maneuverability is pushed to the limit by cramped inner city conditions. Those who need to...
Advance in biomedical imaging: The University of Würzburg's Biocenter has enhanced fluorescence microscopy to label and visualise up to nine different cell structures simultaneously.
Fluorescence microscopy allows researchers to visualise biomolecules in cells. They label the molecules using fluorescent probes, excite them with light and...
NASA's follow-on to the successful ICESat mission will employ a never-before-flown technique for determining the topography of ice sheets and the thickness of sea ice, but that won't be the only first for this mission.
Slated for launch in 2018, NASA's Ice, Cloud and land Elevation Satellite-2 (ICESat-2) also will carry a 3-D printed part made of polyetherketoneketone (PEKK),...
In the last decades, sea level has been rising continuously – about 3.3 mm per year. For reef islands such as the Maldives or the Marshall Islands a sinister picture is being painted evoking the demise of the island states and their cultures. Are the effects of sea-level rise already noticeable on reef islands? Scientists from the ZMT have now answered this question for the Takuu Atoll, a group of Pacific islands, located northeast of Papua New Guinea.
In the last decades, sea level has been rising continuously – about 3.3 mm per year. For reef islands such as the Maldives or the Marshall Islands a sinister...
The ‘Internet of Things’ is growing rapidly. Mobile phones, washing machines and the milk bottle in the fridge: the idea is that minicomputers connected to these will be able to process information, receive and send data. This requires electrical power. Transistors that are capable of switching information with a single electron use far less power than field effect transistors that are commonly used in computers. However, these innovative electronic switches do not yet work at room temperature. Scientists working on the new EU research project ‘Ions4Set’ intend to change this. The program will be launched on February 1. It is coordinated by the Helmholtz-Zentrum Dresden-Rossendorf (HZDR).
“Billions of tiny computers will in future communicate with each other via the Internet or locally. Yet power consumption currently remains a great obstacle”,...
02.02.2016 | Event News
26.01.2016 | Event News
26.01.2016 | Event News
05.02.2016 | Life Sciences
05.02.2016 | Materials Sciences
05.02.2016 | Physics and Astronomy