Inspired by nothing less than the human brain, IBM announced today a new "software ecosystem" that is designed for programming silicon chips that have a dramatically different architecture. The breakthrough technology, the company stated in a press release, could enable “a new generation of intelligent sensor networks that mimic the brain’s abilities for perception, action, and cognition.”

The technology is part of work commissioned by the U.S. Defense Advanced Research Projects Agency (DARPA) for its experimental SyNAPSE (Systems of Neuromorphic Adaptive Plastic Scalable Electronics) project, which was launched in 2009. DARPA has rewarded IBM and its collaborators, which include Cornell University and iniLabs, Ltd., about $12 million in new funding for continuation of the SyNAPSE project; total funding now runs to about $53 million.

“Architectures and programs are closely intertwined and a new architecture necessitates a new programming paradigm,” said Dr. Dharmendra S. Modha, Principal Investigator and Senior Manager, IBM Research.

IBM stated that the goal of SyNAPSE is to “create a system that not only analyzes complex information from multiple sensory modalities at once, but also dynamically rewires itself as it interacts with its environment — all while rivaling the brain’s compact size and low power usage.”

Features of the New Ecosystem

DARPA was established in 1958 “to prevent strategic surprise from negatively impacting U.S. national security and create strategic surprise for U.S. adversaries by maintaining the technological superiority of the U.S. military,” according to its website. IBMs new programming ecosystem is designed for what it refers to as “cognitive computing architectures.”

In 2011, IBM first revealed its brain-inspired chip architecture that is based on a configurable network of “neurosynaptic cores.” The company describes each core as bringing memory (“synapses”), processors (“neurons”), and communication (“axons”) within close proximity in order to execute activity in an event-driven fashion — just like the human brain.

To enable its newly created model, IBM developed support technologies within the programming cycle.

These include a software simulator “of a cognitive computing architecture comprising a network of neurosynaptic cores.” A neuron model forms “a fundamental information processing unit of brain-like computation;” a network of such neurons will be able to sense, remember, and act in response to environmental stimuli in a manner similar to the human brain. Its new programming model is based on reusable building blocks called ‘corelets,’ each of which represents a “blueprint of a network of neurosynaptic cores that specifies a based-level function. Corelets can be combined to produce new corelets that are larger, more complex, or have added functionality.” IBM has also built a program ‘library’ which stores over 150 corelets. Finally, IBM developed a teaching curriculum, referred to as a laboratory, encompassing the “architecture, neuron specification, chip simulator, programming language, application library, and prototype design models.”

And All This Means...?

The title of an article about SyNAPSE published by U.S. Naval Institute News basically spells out a possible goal of the program: "Opinion: Thinking Weapons Are Closer Than We Think." Lieutenant Matthew Hipple writes that the project "signals a new level for biomimicry in engineering."

IBM, though, has long range goals of another order; the company is hoping to build "a chip system with ten billion neurons and hundred trillion synapses, while consuming merely one kilowatt of power and occupying less than two liters of volume." Such a chip system could be embedded within all manner of products from smartphones to medical devices.