Thursday, September 27, 2018

Turing Test : The examination in AI

Turing test, the way to test the knowledge of machine :

          In 1950 a mathematician introduced an examination system to examine the learning process of the machine. His name was Alan Turing. Respecting his invention the method or test was named as Turing effect or Turing test.

       It is like an examination that we sit to pass our classes. There are some questions, that we have to answer. And depending on that answers the examiner gives us marks which state our knowledge and capabilities.

      The same in the case of machines to judge it's ability. How you can say a machine is artificially intelligent! Or what are the lines that after crossing this, a machine can be said to be an AI machine? It is the Turing test. Through this examination process researchers are able to understand a machine's human-like behavior. With the help of Turing test, you can find out a machine artificially intelligent or not!

How this Turing test is conducted :

      Turing test is conducted in a very simple method like it can be seen in the below image.

       As the above picture in the Turing test, there are 3 parts. 2 men and a computer or machine. The examiner asks questions to both the computer and the man. This process goes on and wherever in the middle of the examination. If the computer or the machine is able to make fool the examiner and if the examiner misses just the machine with the man, then the machine or computer can be defined as artificially intelligent.

         This is what a Turing test in a simple manner.

Turing test, turing effect, AI

Types or variations in Turing test:

      There are many versions in the Turing test. These are-

1. Limitation game.
2. Standard interpretation.
3. Limitation game vs Standard tuning test

Variations in the Turing test :

1. Reverse Turing test and captcha.
2. Subject matter expert Turing test.
3. Electronic health records.
4. Hutter prize.
5. Minimum intelligent signal test.
6. Total Turing test.
7. Ebert test.

Conclusion :

      Turing tests are conducted to determine a machine's capability and knowledge. But there is not a full path to examine its intelligence. As we know there is a huge difference between knowledge and intelligence.


Post a Comment