An Equation for Intelligence

Friday, December 6, 2013

 Artificial Intelligence
Explaining how intelligence may be an emergent property rooted in the urge to take control of all possible futures, Alex Wissner-Gross has developed an equation that may explain intelligence itself, and in doing so, may dramatically impact the world of AI.

Richard Feynman once wrote that if human civilization were destroyed and you could pass on only a single concept to our descendants to help them rebuild civilization, that concept should be that all matter around us is made out of tiny elements that attract each other when they are far apart, but repel each other when they are close together.

For Alex Wissner-Gross, the equivalent statement to pass on to descendants to help them build artificial intelligence or to help them understand human intelligence is "That it should be viewed as a physical process that tries to maximize future freedom of action and to avoid constraints in its own future."

This grand view of intelligence has emerged as a guiding principle for Wissner-Gross's own work in development of Entropica, an artificial intelligence platform that has been tested for use in robotic motion planning, computer game play and financial market trading maximization.

Wissner-Gross began his work in artificial intelligence by first trying to understand the fundamental physical mechanisms that underlay intelligence.

In a paper published earlier this year, Wissner-Gross detailed how intelligent behavior stems from the impulse to seize control of future events in the environment. This is the exact opposite of the classic science-fiction scenario in which computers or robots become intelligent, then set their sights on taking over the world.

Related articles
The work led Wissner-Gross to also develop a single equation to explain intelligence. Wissner-Gross calls it, "the closest thing to an E=MC2 for intelligence."  In the equation, F is a force that acts so to maximize future freedom of action, to keep options open with some strength, T with S, an amount of diversity to possible accessible futures, up to some future time horizon, Tau.

In the video above, Wissner-Gross shows how the equation can be applied to artificial intelligence. As the demonstration of Entropica suggests, intelligent behavior doesn't just correlate with the production of long term control of future events, or entropy, it actually emerges from it.

With the equation as a fundamental algorithm, Entropica was able to pass multiple animal intelligence tests, play games and even earn money trading stocks, all without being instructed to do so.

Wissner-Gross is an American scientist, inventor, and entrepreneur. In 2003, he became the last person in MIT history to receive a triple major, with bachelors degrees in physics, electrical engineering, and mathematics, while graduating first in his class from the MIT School of Engineering. In 2007, he completed his Ph.D. in Physics at Harvard, where his research on programmable matter, ubiquitous computing, and machine learning was awarded the Hertz Doctoral Thesis Prize. He currently holds academic appointments as an Institute Fellow at the Harvard University Institute for Applied Computational Science and as a Research Affiliate at the MIT Media Lab.


By 33rd SquareSubscribe to 33rd Square