This article is cited in 1 paper

Part 2. Mathematics and Computer Science

Optimization of neural network learning via system with automata approach

V. A. Biryukova

Lomonosov Moscow State University

Abstract: This paper presents an automata model for learning neural networks, implemented using the high-level programming language python and tested on the problem of binary classification. Also, within the system under consideration, a procedure for automating the neural network training by choosing a training strategy from the point of view of "best practices" is implemented.

Keywords: artificial neural networks, finite automata, non-finite automata, system for working with neural networks, automated machine learning, autoML, hyperparameter optimization via best practices, computer vision, CV, binary classification.