Facebook has taught its AI to solve the most complex math problems. This was achieved thanks to the fact that the neural network began to perceive tasks linguistically - that is, as a kind of language similar to Russian or English.

According to hightech.fm, existing neural networks operate on the basis of the so-called approximation system: they are able to recognize objects in an image or text (for example, dogs, letters, balls, etc.), compose an equation from this data and solve it.

These operations require a lot of time and, as a result, significant computing power. Facebook engineers found a solution to this problem - they taught their neural network to consider the problem not as a mathematical one, but rather as a language problem.

In other words, the researchers taught their AI to speak a mathematical language. The result of this approach was the acceleration of problem solving by several orders of magnitude compared with systems for solving algebraic problems such as Maple, Mathematica, and Matlab.

© All rights reserved. Citing to www.ict.az is necessary upon using news