A Novel Fractional Gradient-Based Learning Algorithm for Recurrent Neural Networks

作者：Khan, Shujaat; Ahmad, Jawwad; Naseem, Imran^*; Moinuddin, Muhammad

来源：Circuits, Systems, and Signal Processing, 2018, 37(2): 593-612.

DOI：10.1007/s00034-017-0572-z

• Back-propagation through time (BPTT)
• Recurrent neural network (RNN)
• Gradient descent
• Fractional calculus
• Mackey-Glass chaotic time series
• Minimum redundancy and maximum relevance (mRMR)

In this research, we propose a novel algorithm for learning of the recurrent neural networks called as the fractional back-propagation through time (FBPTT). Considering the potential of the fractional calculus, we propose to use the fractional calculus-based gradient descent method to derive the FBPTT algorithm. The proposed FBPTT method is shown to outperform the conventional back-propagation through time algorithm on three major problems of estimation namely nonlinear system identification, pattern classification and Mackey-Glass chaotic time series prediction.

• 出版日期2018-2