Researchers in the field of neuromorphic computing at MIT(Massachusetts Institute of Technology) have attempted to design a neuromorphic chip that works like human brain.
Let us first see how the human brain works – The neurons in our brain sends signals to other neurons via synapse (space between the cells). The synapse works as a bridge for the nerve cells to communicate. There are about 100 billion neurons in the brain, each firing off 5-50 messages per second. This activity allows us to process our environment, move our muscles, make conscious movements, make decisions, form memories, etc.
A team of researchers at the MIT university have attempted to design an artificial synapse that can precisely control the strength of an electric current flowing across it, similar to the way ions flow between neurons in the brain.
The researchers fabricated a neuromorphic chip consisting of these artificial synapses made from silicon germanium, each synapse measuring about 25 nanometers across. They applied voltage to each synapse and found that all synapses exhibited more or less the same current, or flow of ions, with about a 4 per cent variation between synapses. This ‘brain-on’ chip introduced, works on analogue fashion (unlike the previously fabricated computer chips which worked on digital signals) so that it can process multiple parallel computations, like our brain does.
The team tested a single synapse over multiple trials, applying the same voltage over 700 cycles, and found the synapse exhibited the same current, with just 1 per cent variation from cycle to cycle. In simulations, they found that the chip and its synapses could be used to recognize samples of handwriting, with 95 per cent accuracy.
“Ultimately we want a chip as big as a fingernail to replace one big supercomputer,” Kim says, leader of the research team at MIT.
To read more, visit here.
Our take on this:
It can prove to be a major breakthrough towards portable Artificial Intelligence devices (smartphones, laptops etc). These neuromorphic chips would make it easier for us to enable the devices with AI technology, which will be faster, more efficient and use less power.
An avid reader and blogger who loves exploring the endless world of data science and artificial intelligence. Fascinated by the limitless applications of ML and AI; eager to learn and discover the depths of data science.
Hi, Really this is really an amazing Neuromorphic Chip that was introduced by MIT , but is it helpful for the ANN and other ML Algorithms internally ?
Yes of course! Take an example of ASICs (https://en.wikipedia.org/wiki/Application-specific_integrated_circuit), which are circuits designed specifically for a task. These ASICs have a great boost over general purpose chips when we compare them. Similarly, neuromorphic chips are specifically designed so that ANNs and ML algorithms would work faster and more efficiently.