We communicate with people, understand writing, or perceive objects through sight. That is, our brain solves these “problems,” but we cannot write down what rules it follows. For example, we quickly define a cat’s concept in a variety of images, even if they are of different sizes, colors, shapes, or if it is a toy cat. A person understands that in front of him is a cat, even if he/she sees only its eye.

We are faced with a fundamental limitation of the solution to the problem when we cannot identify explicit rules that the algorithm should follow. And we need a new approach to automation, different from the classical one. This is where artificial intelligence comes.

AI bypasses the error of the classical approach. This allows the system to identify patterns and learn implicit rules (usually by analyzing thousands of examples) to analyze data (images, sound files, texts, etc.) according to specific schemes (as was the case with the cat).

Read More…