The research aims to conduct neural network modeling of the process of innovative development of the radio-electronic industry (REI) in Russian regions. This makes it possible to identify regions leading in this industry and regions with prospects for the future development of the electronic energy industry within their territory. The authors developed a methodology for neural network modeling of the innovative development of the regional radio-electronic industry, which includes the following stages: (1) collection, adjustment for inflation, and standardization of the necessary data; (2) construction and training of a neural network for the regression problem; (3) verification of the trained network on data from a new observation period: (4) identification of leading regions and segments of planned network inputs; (5) assessment of the compliance of regions with planned network inputs. In this case, the target functions are as follows: (1) the volume of innovative goods (total), (2) developed advanced production technologies (total), (3) balanced financial result (informatization and communications), and (4) export of technologies (receipt of funds). In the Matlab program, a neural network was built and trained for a quasi-time series from 2010 to 2020 for 83 regions of Russia. The correlation coefficients of the actual target data with those obtained in the model are close to 1. The error histogram is close to the normal distribution law. For goal 2, the average relative model error (MAPE) on the new 2021 data is 61.2%. A comparison of the target and predicted value graphs for all four target functions illustrates the good quality of the network prediction. The results obtained may be useful to government agencies for planning support for the innovative development of the electronic energy industry in these regions. These results can be used by investors to select directions for capital investments of their funds.