Who invented the Turing test? Turing test questions

Probably, today not such person who at least once did not hear about such concept, as the test of Alan Turing. Probably, the majority, in general, is far from understanding what such a testing system is. Therefore, we will dwell on it in some detail.

What is a Turing test: the basic concept

Back in the late 40-ies of the last century, a lot of scientists were studying the problems of the first computer development. It was then that one of the members of some non-governmental group Ratio Club, engaged in research in the field of cybernetics, asked a perfectly logical question: Is it possible to create a machine that would think like a person, or at least imitate his behavior?

Do I need to say who invented the Turing test? Apparently not. For the initial basis of the whole concept, which is still relevant, the following principle has been taken: will a person, for some time, communicate with an invisible interlocutor on completely different arbitrary topics, determine who is a real person or machine in front of him? In other words, the question is not only to simulate the machine behavior of a real person, but also to find out whether she can think for herself. No doubt, this issue remains controversial until now.

History of creation

In general, if we consider the Turing test as an empirical system for determining the "human" capabilities of a computer, it should be said that the indirect basis for its creation was the curious statements of the philosopher Alfred Ayer, which he formulated back in 1936.

Sam Ayer compared, so to speak, the life experience of different people, and on the basis of this expressed the view that the soulless machine will not be able to pass any test, because it can not think. At best, it will be pure imitation water.

In principle, so it is. To create a thinking machine, one imitation is not enough. Very many scientists as an example lead the Wright brothers who built the first plane, abandoning the tendency to imitate birds, which, by the way, was peculiar to a genius such as Leonardo da Vinci.

Istria does not say whether Alan Turing himself (1912-1954) himself knew about these postulates, nevertheless in 1950 he compiled a whole system of questions that could determine the degree of "humanization" of the machine. And I must say, this development is still one of the fundamental, though, already when testing, for example, computer bots, etc. In reality, the principle turned out to be such that only a few programs managed to pass the Turing test. And then, "pass" - said with great reserve, because the result of testing has never had a rate of 100 percent, at best - just over 50.

At the very beginning of his studies, the scientist used his own invention. It was called the "Turing test machine." Since all conversations were supposed to be introduced exclusively in printed form, the scientist asked several basic directives on writing answers, such as moving the printed tape left or right, printing a certain symbol, etc.

ELIZA and PARRY programs

Over time, programs began to become more complex, and two of them, in situations where the Turing test was applied, showed striking results at the time. These became ELIZA and PARRY.

As for "Eliza", created in 1960: on the basis of the question, the machine had to determine the key word and on its basis make a reverse answer. It was this that made it possible to deceive real people. If there was no such word, the machine returned a generalized answer or repeated one of the previous ones. However, the passage of the "Eliza" test is still in doubt, since the real people who communicated with the program were initially trained psychologically in such a way that they thought in advance that they were talking to a person, not a car.

The PARRY program is somewhat similar to Eliza, but was created to simulate paranoid communication. What is most interesting, real patients of clinics were used for its testing. After recording transcripts of conversations in the teletype mode, they were evaluated by professional psychiatrists. Only in 48 percent of cases they were able to correctly assess where the person is, and where the machine is.

In addition, almost all of the then-running programs worked in the light of a certain period of time, since people in those days thought much faster than the machine. Now - on the contrary.

Supercomputers Deep Blue and Watson

Interesting enough to look at the development of IBM, which is not something to think, but had incredible processing power.

Probably, many remember how in 1997 a supercomputer Deep Blue won 6 chess games against then-current world champion Garry Kasparov. Actually, the Turing test is applicable to this machine very conditionally. The thing is that it initially was set a lot of templates of parties with an incredible amount of interpretation of the development of events. The car could estimate the order of 200 million positions of figures on a board in a second!

Computer Watson, consisting of 360 processors and 90 servers, won the American game show, having bypassed all the parameters of the other two participants, for which, in fact, received a $ 1 million prize. Again, the question is controversial, since incredible volumes of encyclopedic data were put into the machine, and the machine simply analyzed the question for the presence of a keyword, synonyms or generalized coincidences, and then gave the correct answer.

Emulator Eugene Goostman

One of the most interesting events in this area was the program of Odessa Eugene Gustman and Russian engineer Vladimir Veselov, now living in the US, who imitated the identity of a 13-year-old boy.

On June 7, 2014, the Eugene program showed its full potential. Interestingly, 5 bots and 30 real people took part in the testing. Only in 33% of cases out of a hundred juries could determine that this is a computer. The point here is that the task was complicated by the fact that the child's intelligence is lower than that of an adult person, and less knowledge.

The questions of the Turing test were the most general, however, for Eugene (Euegene) there were some specific questions about the events in Odessa, which could not be overlooked by any inhabitant. But the answers still made me think that the jury was a child. So, for example, the program answered the question of residence right away. The code was asked whether the interlocutor of this or that number was in the city, the program stated that he did not want to talk about it. When the interlocutor tried to insist on a conversation in line with what exactly happened on that day, Eugene disowned himself by saying that they should know for themselves what to ask him? In general, the child's emulator turned out to be extremely successful.

Nevertheless, it's still an emulator, not a thinking creature. So the uprising of the machines will not take very long.

but on the other hand

Finally it remains to add that so far there are no prerequisites for creating thinking machines in the near future. Nevertheless, if earlier recognition issues were related to machines, now that you are not a machine, you have to prove to almost every one of us. Look at least to input captcha on the Internet to gain access to some action. While it is believed that no electronic device has been created yet, capable of recognizing a twisted text or a set of symbols, except for a person. But who knows, everything is possible ...