S.D. Ionov - Post-graduate Student, Institute of Mathematics and Mechanics (IMM), Ural Branch of the Russian Academy of Sciences. E-mail: progsdi@gmail.com

In the article «Self-reproducing artificial intelligent neural networks language (SAINNL)» the descriptive language of a specific type of neural networks called the self-reproducing artificial intelligent neural networks (SAINN) is presented. This article is divided into 7 parts. In the beginning of the article the author defines the requirements for the language and provides a review of the existing analogs. Next a general description of the SAINNL elements is presented. The third part provides the features of static and dynamic generation, which are the basis of growth functions during the process of compilation and at runtime. Useful for developers string definitions and commentaries are in the fourth and fifth parts accordingly. In the sixth part based on the proposed main components the examples of SAINNL application for description of separate blocks, as well as, fully functional networks and in particular perceptron are given. In conclusion, the summary on the built SAINNL is made, and it is stated that it can be applied in development of algorithms based on SAINN.