Researchers at the Massachusetts Institute of TechnologyMassachusetts Institute of Technology) claim to have created artificial intelligence a million times faster than the human brain. In the manufacturing process, they have used a new inorganic material that can provide extreme speeds and superior energy efficiency.

With the constant development of machine learning, training more complex neural network models requires more and more time, energy and money. The emerging solution to this is what they call analog deep learning, which promises faster computations with negligible power consumption. As MIT explains:

“Programmable resistors are the basic building blocks of analog deep learning. […]. By iterating arrays of programmable resistors in complex layers, researchers can create an analog network of artificial “neurons” and “synapses” that perform calculations similar to a digital neural network. This network can be trained to perform complex AI tasks such as image recognition and natural language processing.”

“Programmable resistors are the basic building blocks of analog deep learning. […]. By iterating arrays of programmable resistors in complex layers, researchers can create an analog network of artificial “neurons” and “synapses” that perform calculations similar to a digital neural network. This network can be trained to perform complex AI tasks such as image recognition and natural language processing.”

These programmable resistors greatly increase the speed of training the neural network while greatly reducing the cost and energy to perform this training.

In the future, people will be indistinguishable from machines.
Evolution or destruction? The fate of mankind may be in the hands of artificial intelligence

A recent MIT creation, in turn, is based on analog synapses, which supposedly outnumber the synapses in our brain. The key element of the new technology is known as proton programmable resistor. Researchers have replaced organic media with inorganic phosphosilicate glass (PSG), mostly silicon dioxide, resulting in nanosecond speeds. According to Joo Lee, Senior Author and Professor of Nuclear Science:

“The action potential in biological cells rises and falls on a time scale of milliseconds as a potential difference of around 0.1 volts is limited by the stability of water. Here we apply up to ten volts to the glass film. a solid that conducts protons without permanently damaging it. And the stronger the field, the faster the ion devices work.”

“The action potential in biological cells rises and falls on a time scale of milliseconds as a potential difference of around 0.1 volts is limited by the stability of water. Here we apply up to ten volts to the glass film. a solid that conducts protons without permanently damaging it. And the stronger the field, the faster the ion devices work.”

Because inorganic phosphosilicate glass can withstand high loads without breaking, it allows protons to move at absurd speeds and is also energy efficient. Another important point is that the material is common and easy to manufacture.

Research published in the journal the science. Going forward, the researchers plan to upgrade these programmable resistors for high-volume production, as well as study materials to further eliminate the bottlenecks that limit the voltage required to transport protons “in, through, and out of the electrolyte.”

According to Jesús A. del Alamo, another study author and professor in the Department of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology (EECS), “Our collaboration will be essential to innovation in the future. The road ahead will continue to be very challenging, but at the same time very exciting.”

Via: Futurism, MIT, TweakTown

…..