Neural Networks and Classical Algorithms in Inverse Problems of Anomalous Diffusion - Доклады на конференциях | Sciact - Система учета научной деятельности ИМ СО РАН

Neural Networks and Classical Algorithms in Inverse Problems of Anomalous Diffusion Доклады на конференциях

Язык

Английский

Тип доклада

Секционный

Конференция

Science and Artificial Intelligence Conference, S.A.I.ence
14-15 нояб. 2020 , Новосибирск

Авторы

Dedok V.A. ¹ , Bugueva T.V. ¹

Организации

1	Институт математики им. С.Л. Соболева СО РАН

The paper develops a new numerical method for inverse problem solving. This method can be classified as a predictor-corrector method, in which an artificial neural network plays the role of a predictor, and a gradient method plays the role of a corrector. We apply this method to inverse anomalous diffusion problem and show its statistical efficiency. Index Terms—inverse problems, anomalous diffusion, artificial neural networks, numerical methods.

Dedok V.A. , Bugueva T.V.
Neural Networks and Classical Algorithms in Inverse Problems of Anomalous Diffusion
Science and Artificial Intelligence Conference, S.A.I.ence 14-15 Nov 2020