Researchers "taught" artificial intelligence to create new colors

Artificial intelligence has made tremendous progress and is completely superior in terms of human abilities. Fortunately, there are many areas in which carbon breaks silicon, and one of them is to create flowers.

Computer scientist Janelle Shane developed an algorithm of machine learning that can not only create colors, but also call them themselves. It is possible to note with pleasure, that creative people in this case do not threaten anything. They are in complete safety from the rebellion of robots, guessing colors.

Artificial neural network

"Considering the structure of an artificial neural network as a whole, we can conclude that it really takes a greater degree of brown, beige and gray colors. Artificial intelligence really has a very poor ability to determine the different shades of colors, "wrote scientist Shane in a message on his page Tumblr.

Among the inventions of artificial intelligence there is a bright laureate named Gray Pubic, a thin pink called Bank Butt, a greenish gray, known as Snowbonk, and an antique pink that the computer called Testing. But a special achievement is sandy-brown, which was simply called Turdly ("severe" in English).

Although the names seem strange, but the work of scientists over them was very fascinating. The starting point was the Sherwin-Williams catalog, which includes 7,700 colors.

Smart algorithm

As explained by Shane Ars Technica, he used an algorithm that can guess all the symbols in a certain sequence. It is very interesting to create new machines based on RGB values that can determine how much red, green and blue in the combined color.

"Later in the educational process, the artificial neural network will be trained even better. - continues Shane in his blog. - By this moment AI can determine such basic colors as white, red and gray, although not always unmistakably. "

Supply with necessary information

At present, computer scientists and programmers are testing the limits of the possibilities of machine learning. The key to an innovative approach is to give computers the opportunity to learn. Algorithms are not programmed to understand colors, but machines can get a general idea if the inventors provide them with sufficient information.

The computer science allowed programming AlphaGo, which last year became the first artificial intelligence. He managed to win a man from the party in the ancient game Go. The machine is also used in online recommendations, in the detection of fraud and even for the purpose of training machines with self-management.