Deep inference: A convolutional neural networks method for parameter recovery of the fractional dynamics

نویسندگان
Biranvand, Nader ID 1
Khalili, Ali ID 2
Amani Rad, Jamal ID 3
Hadian-Rasanan, Amir Hossein ID 4

1 Faculty of Sciences, Imam Ali University, Tehran, Iran

2 Faculty of Engineering, Imam Ali University, Tehran Iran

3 Department of Cognitive Modeling, Institute for Cognitive and Brain Sciences, Shahid Beheshti University, Tehran, Iran

4 Department of Cognitive Modeling, Institute for Cognitive and Brain Sciences, Shahid Beheshti University, Tehran, Iran

doi
10.22075/ijnaa.2021.4757
چکیده

Parameter recovery of dynamical systems has attracted much attention in recent years. The proposed methods for this purpose can not be used in real-time applications. Besides, little works have been done on the parameter recovery of the fractional dynamics. Therefore, in this paper, a convolutional neural network is proposed for parameter recovery of the fractional dynamics. The presented network can also estimate the uncertainty of the parameter estimation and has perfect robustness for real-time applications.

کلیدواژه‌ها
Convolutional Neural Network, Parameter estimation, Fractional Dynamics, Data driven discove