Интеллектуальные системы в науке и технике. Искусственный интеллект в решении актуальных социальных и экономических проблем ХХI века
The collection contains materials from the international conference "Intelligent systems in science and technology" and the Sixth all-Russian scientific and practical conference "Artificial intelligence in solving urgent social and economic problems of the XXI century", which was held on October 12-18, 2020 in Perm as part of the Perm natural science forum "Mathematics and global challenges of the XXI century".
The collection is intended for researchers and teachers, teachers, postgraduates, undergraduates, students and anyone interested in the problems and prospects of development and application of artificial intelligence methods
The article describes the development of the music generation system based on a genetic algorithm which is used with an artificial neural network as the corresponding fitness function. The application allows you to create melodies which are based on initially set melodies of a certain musical genre selected by the user, which are modified and evaluated by a neural network during the execution of a genetic algorithm to determine whether they belong to the genre specified by the user. Using a genetic algorithm, the fitness function of which is represented by a trained neural network that solves the problem of music classification, allows you to automate the process of generating melodies.
The conclusion is confirmed that there is no unified approach to solving the issue of civil liability for the actions of artificial intelligence. The authors proposed the use of guilty liability, liability regardless of guilt and liability, based on a risk-oriented approach. For the purpose of determining the type of liability for the action of AI, the classification of AI technologies is proposed on four grounds: work in changing environmental conditions; work in conditions of the appearance of new points; the number of tasks to be solved and functions performed, recording information about the activities of artificial intelligence technology. According to the authors, the most promising approach to the legal regulation of liability for the functioning of artificial intelligence technologies is a risk-oriented approach. The presented conclusions testify to the relevance of the research topic and the beginning of the formation of a scientific understanding of civil liability for the actions of AI in Russia and abroad