Abstract
In this paper, we propose a novel computational architecture of memristor-based echo state network (MESN) with the online least mean square (LMS) algorithm. Newman and Watts small-world network is adopted for the topological structure of MESN network with memristive neural synapses. In the MESN network, the state matrix of the reservoir layer, which is obtained by raising the dimension of input data, is utilized as an input of the LMS algorithm to train the output weight matrix on chip. After certain iterations, the resistance value of memristor is adjusted to a constant. Thus, the final weight output matrix is obtained. To verify the effectiveness of the proposed MESN network, car evaluation and short-term power load forecasting are employed with the effect evaluation of the node number and the connectivity degree of the reservoir layer. The research provides a novel way to design neuromorphic computing systems.
| Original language | English |
|---|---|
| Article number | 8354924 |
| Pages (from-to) | 1787-1796 |
| Number of pages | 10 |
| Journal | IEEE Transactions on Systems, Man, and Cybernetics: Systems |
| Volume | 49 |
| Issue number | 9 |
| Early online date | 4 May 2018 |
| DOIs | |
| Publication status | Published - Sept 2019 |
| Externally published | Yes |
Bibliographical note
Publisher Copyright:© 2018 IEEE.
Funding
This work was supported in part by the Natural Science Foundation of China under Grant 61673187, Grant 61673188, and Grant 61773081, in part by the Technology Transformation Program of Chongqing Higher Education University under Grant KJZH17102, and in part by NPRP from the Qatar National Research Fund (a member of Qatar Foundation) under Grant NPRP 9-466-1-103.
Keywords
- Echo state network
- least mean quare
- memristor
- neural network