Cognitive computing: IBM unveils software for its brain-like SyNAPSE chips
From the firm that brought you the 'cat brain chip'
IBM has unveiled a new computing language to help boffins better program the mad scientist brain-like silicon chips first envisaged in the first three phases of what is known at The Reg as DARPA's "mad cat brain chip" project.
Prototypes of the chips with the cool cognitive computing architectures have already been unveiled to the world, but Big blue says its new programming model will take the complexity of the project to a new level.
According to a Big Blue press release, the technology can mimic the "function, low power, and compact volume of the brain".
It could be used to create next-generation intelligent sensor networks which are capable of tricks like perception, action and cognition - the sort of things humans take for granted. This could allow for human-like collection and analysis of masses of big data, akin to that performed by our human eyes and brains while we navigate the world.
"Architectures and programs are closely intertwined and a new architecture necessitates a new programming paradigm," said Dr Dharmendra Modha, principal investigator and senior manager at IBM Research. "While complementing today's computers, this will bring forth a fundamentally new technological capability in terms of programming and applying emerging learning systems."
IBM's new programming model is "dramatically different" from traditional stored-program architectures, say the boffins.
Big Blue's latest wheeze dispenses with the sequential operation inherent in von Neumann architectures, replacing it with a "distributed, highly interconnected, asynchronous, parallel, large-scale cognitive computing architectures".
Modern computers are superb number-crunchers, but often find it difficult to cope with "real-time processing of the noisy, analog, voluminous, Big Data produced by the world around us," IBM's researchers continued.
The brain, which operates relatively slowly in calculations and offers a low precision, is nonetheless excellent at sifting through staggering amounts of data, before recognising and acting upon patterns.
Impressively, while interpreting the real world, the brain “uses the same amount of power as a 20-watt light bulb.”
IBM first unveiled a chip based on the brain in 2011. The IBM team created a chip that is designed to chew on streams telemetry and rewire itself, much as your brain does as it learns, as it learns about the world from that telemetry.
So far IBM's effort to build a brain has cost about $53m, with $2m in new funding recently pledged by the US Defense Advanced Research Projects Agency (DARPA), the US Army's battalion of crazed weapon inventors.
The project is known as Systems of Neuromorphic Adaptive Plastic Scalable Electronics (SyNAPSE).
In the long term, IBM wants to build a system with ten billion neurons and hundred trillion synapses, which uses just one kilowatt of power and occupies less than two litres in volume.
In a statement, IBM said:
Systems built from these chips could bring the real-time capture and analysis of various types of data closer to the point of collection. They would not only gather symbolic data, which is fixed text or digital information, but also gather sub-symbolic data, which is sensory based and whose values change continuously. This raw data reflects activity in the world of every kind ranging from commerce, social, logistics, location, movement, and environmental conditions.
The technology could be used to copy the human eye, IBM claimed, each of our working eyes deals with more than a terabyte of information each day. Big Blue suggested its chips could build visors for blind people which capture information and then allow blind people to navigate around the world, following instructions conveyed over headphones.
The initial phase of the project simulated the cortex of a cat brain on an IBM BlueGene massively parallel supercomputer with 147,456 cores and 144TB of memory developing the basic synaptic circuits for the brain chip. ®
Sponsored: Beyond the Data Frontier