Mapping the (Neural) Path Towards Cognitive Computing

By November 18, 2009

A team of IBM and Berkeley scientists is trying to build a computing system based on the architecture of the human brain, in an effort to fundamentally change the way computers function. The research attempts to simulate the brain’s abilities to sense, perceive, act, interact and think, as well as its low power consumption and compact size. “As the digital and physical worlds continue to merge and computing becomes more embedded in the fabric of our daily lives, it’s imperative that we create a more intelligent computing system that can help us make sense the vast amount of information that's increasingly available to us, much the way our brains can quickly interpret and act on complex tasks,” said Josephine Cheng, IBM Fellow and lab director of IBM Research - Almaden, which heads the research.

The scientists believe that cognitive computing will better meet the needs of future users than traditional von Neumann computing, which is focused on storage, memory and a computation/data dichotomy.

Cognitive computing, on the other hand, will replicate neurons and synapses, “blurring the boundary between computation and data,” according to the press release.

Cognitive computing will enable computers to react quickly to disparate pieces of information and process them based on context and past experience, spot patterns in random data, and deal with ambiguity, like the human brain.

To do this, the brain's underlying architecture needs to be better understood.

The scientists combined advanced brain mapping with a new algorithm to synthesize neurological data. The BlueMatter algorithm maps connections between cortical and sub-cortical regions in the brain, based on a near real-time cortical brain simulation.

“Mapping the wiring diagram of the brain is crucial to untangling its vast communication network and understanding how it represents and processes information,” according to the press release.

It's oddly appropriate that the research is headed by IBM, as the company's namesake, 2001's HAL -- 'IBM' with every letter shifted one place back --  is the most famous sentient computer in pop-culture history. And we know how well that went.

Let's hope for better results this time, Dave.

UPDATE: (Nov 19, 2:17 PST). The scientists have announced that their simulation level has surpassed the level of complexity of a cat's cortex. The similation now contains the equivalent of one billion neurons and ten trillion synapses.

Legal mentions © L’Atelier BNP Paribas