Analog Neural Circuit with Switched Capacitor and Design of Deep Learning Model

In the neural network field, many application models have been proposed. Previous analog neural network models were composed of the operational amplifier and fixed resistance. It is difficult to change the connecting weight of network. In this study, we used analog electronic multiple and sample hold circuits. The connecting weights describe the input voltage. It is easy to change the connection coefficient. This model works only on analog electronic circuits. It can finish the learning process in a very short time and this model will enable more flexible learning. However, the structure of this model is only one input and one output network. We improved the number of unit and network layer. Moreover, we suggest the possibility of realization about the hardware implementation of the deep learning model.

[1]  Naohiro Ishii,et al.  Analog Learning Neural Network Using Multiple and Sample Hold Circuits , 2012, 2012 IEEE/ACIS 11th International Conference on Computer and Information Science.

[2]  道木 慎二,et al.  Realization of Back Propagation Learning for Pulsed Neural Networks Based on Delta-Sigma Modulation and Its Hardware Implementation , 2005 .

[3]  H. Imamura,et al.  Analysis of a simple A/D converter with a trapping window , 2003, Proceedings of the 2003 International Symposium on Circuits and Systems, 2003. ISCAS '03..

[4]  Naohiro Ishii,et al.  Analog VLSI Layout Design and the Circuit Board Manufacturing of Advanced Image Processing for Artificial Vision Model , 2008, KES.

[5]  Naohiro Ishii,et al.  Dynamic Learning of Neural Network by Analog Electronic Circuits , 2011, KES.

[6]  Masashi Kawaguchi,et al.  Motion Detecting Artificial Retina Model by Two-Dimensional Multi-Layered Analog Electronic Circuits , 2003, IEICE Trans. Fundam. Electron. Commun. Comput. Sci..

[7]  C.A.T. Salama,et al.  Image-motion detection using analog VLSI , 1992 .

[8]  Naohiro Ishii,et al.  Analog Real Time Learning Neural Network using Multiple and Sample Hold Circuits , 2012, KES.

[9]  Bertram E. Shi,et al.  Subpixel resolution binocular visual tracking using analog VLSI vision sensors , 2000 .

[10]  M. Kawaguchi,et al.  Analog VLSI layout design of advanced image processing for artificial vision model , 2005, Proceedings of the IEEE International Symposium on Industrial Electronics, 2005. ISIE 2005..

[11]  Carver Mead,et al.  Analog VLSI and neural systems , 1989 .

[12]  Pascal Vincent,et al.  Representation Learning: A Review and New Perspectives , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Daniela Dragomirescu,et al.  A cellular analog network for MRF-based video motion detection , 1999 .