MIT scientists take a step closer to artificial intelligence

Researchers at Massachusetts Institute of Technology (MIT) have developed an analogue silicon chip with 400 transistors that emulates the activity of a brain synapse in the first step to building truly intelligent systems.

Researchers at Massachusetts Institute of Technology (MIT) have developed an analogue silicon chip with 400 transistors that emulates the activity of a brain synapse, which is the connection between two neurons, in the first step to building truly intelligent systems.

There are about 100 billion neurons in the brain, each of which forms synapses with many other neurons. This process is believed to underpin many brain functions, such as learning and memory.

The chip, described in the latest edition of the journal Proceedings of the National Academy of Sciences, will allow neuroscientists to conduct basic research on how the brain actually works and could lead to the study and treatment of diseases related to brain malfunction.

But the chip could also potentially improve devices that allow people to operate things such as computer mice with their thoughts and create artificial intelligence devices that replicate brain behaviour for tasks such as pattern recognition, cognition, learning, memory and decision-making.

 Chi-Sang Poon, a research scientist in the Harvard-MIT Division of Health Sciences and Technology, told MSNBC: "We are not talking about recreating a whole brain at this point. We have to start with one system."  

Unlike digital computer chips that treat the function of neurons like a simple on/off switch, Poon said the MIT chip get into the nitty-gritty of how the neurons work intra-cellularly, which involves all the ionic processes that are going on.

Activity in the synapses relies on so-called ion channels which control the flow of charged atoms such as sodium, potassium and calcium.

The MIT team said understanding how the brain works will enable scientists to reverse-engineer it and put it in a chip to reproduce those functions.

The team plans to use their chip to build systems to model specific neural functions, such as visual processing, according to the BBC.

Such systems could be much faster than computers, and the chip could ultimately prove to be even faster than the biological process, said the researchers.

Read more on IT suppliers

CIO
Security
Networking
Data Center
Data Management
Close