What is artificial intelligence?
What is intelligence? Is it the ability to perceive the world, to predict the immediate or distant future, or to plan a series of actions to achieve a goal? Is it the ability to learn, or the ability to apply knowledge wisely? The definition is difficult to define.
Artificial intelligence (AI) consider as a set of techniques that allow machines to perform tasks and solve problems normally reserved for humans and certain animals. Artificial intelligence, therefore, consists in implementing some techniques designed to enable machines to imitate a form of real intelligence.
History of artificial intelligence:
The concept was born in the 1950s thanks to Alan Turing. In his book Computing Machinery and Intelligence, the latter raises the question of bringing a form of intelligence to machines. He then described a test now known as the “Turing Test” in which a subject interacts blindly with another human, then with a machine programmed to formulate meaningful responses. If the subject is not able to make a difference, then the machine has passed the test and, according to the author, can consider this machine intelligent.
AI tasks are sometimes very simple for humans, such as recognizing and locating objects in an image, planning the movements of a robot to grab an object, or driving a car. They sometimes require complex planning, such as playing chess or Go. The most complicated tasks require a lot of knowledge and common sense, for example translating a text or conducting a dialogue.
In recent years, intelligence has almost always been associated with learning abilities. It is through learning that an intelligent system capable of performing a task can improve its performance with experience. It is through learning that they will be able to learn to perform new tasks and acquire new skills. The field of AI has not always considered learning as essential to intelligence.
In the past, building an intelligent system consisted in writing a program “by hand” to play chess (by tree-search), recognize printed characters (by comparison with prototype images), or make a medical diagnosis based on symptoms (by logical deduction based on rules written by experts). But this “manual” approach has its limits.
The challenges of artificial intelligence today:
The opportunities are such that AI, especially deep learning, is seen as a technology of strategic importance for the future. Advances in computer vision are paving the way for driverless cars and automated medical imaging analysis systems. Already, some high-end cars use the vision system of the Israeli company MobilEye, which uses a convolutional network for driving assistance.
Medical image analysis systems detect melanomas and other tumors more reliably than experienced radiologists. At Facebook, Google and Microsoft, image recognition systems allow the search and organization of photos and the filtering of violent or some other images.
For several years now, all smartphone speech recognition engines have been using deep learning. Considerable R&D efforts are devoted to natural language processing: text comprehension, question-and-answer systems, dialogue systems for virtual agents, and machine translation. In this area, the profound learning revolution has been heralded but is not yet complete. Nevertheless, rapid progress is being made. In the last international machine translation competition, the winner used a recurring network.