궤환성을 갖는 단츰신경회로망의 Inhibitory Synapses

Inhibitotory Synapses of Single-layer Feedback Neural Network

  • 발행 : 2000.11.01

초록

The negative weight can be ofter seen in Hopfield neural network, which is difficult to implement negative conductance in circuits. Usually, the inverted output of amplifier is used to avoid negative resistors for expressing the negative weights in hardware implementation. However, there is some difference between using negative resistor and the inverted output of amplifier for representing the negative weight. This difference is discussed in this paper.

키워드

참고문헌

  1. Hopfield, J. J. 'Neurons with graded response have collective computational properties like those of two state neurons.' Proc. National Academy of Sciences 81: 3088-3092, 1984 https://doi.org/10.1073/pnas.81.10.3088
  2. Hopfield, J. J., and D. W. Tank, 'Neural Computation of Decisions in Optimization Problems,' Biolog. Cybern. Vol. 52, 141-154, 1985 https://doi.org/10.1007/BF00339943
  3. Hopfield, J. J., and D. W. Tank. 1986. 'Computing with Neural Circuits: A model,' Science Vol. 23, 625-633, 1986 https://doi.org/10.1126/science.3755256
  4. J. M. Zurada, 'Gradient-type Neural Systems for Computation and Decision-Making', Progress in Neural Networks, Vol. 2, Ablex Publishing Corp., O. M. Omidvar, Editor, April, 1991
  5. Mengkang Peng, Narebdra K. Gupta and Alistair F. Armitage, 'An investigation into the Improvement of Local Minima of the Hopfield Network' Neural Networks, Vol. 9, pp. 1241-1253, 1996 https://doi.org/10.1016/0893-6080(96)00017-2
  6. Naoki Kurita and Ken-ichi Funahashi, 'On the Hopfield Neural Networks and Mean Field Theory' Neural Networks, Vol. 9, pp. 1531-1540, 1996 https://doi.org/10.1016/0893-6080(95)00129-8
  7. Armelle Le Gall and Vassilis Zissimopoulous, 'Extended Hopfield Models for Combinatorial Optimization', Neural Networks, Vol. 10, No.1, Jan., 1999 https://doi.org/10.1109/72.737495
  8. Min J. Kang, 'Input Conductance of Neuron for Hopfield type Neural Networks', submitted to IEEE Korea Council, 2000
  9. Eugene M. Izhikaevich, 'Class 1 Neural Excitability, Conventional Synapses Weakly Connected Networks, and Mathematical Foundations of Pulse-Coupled Models', Neural Networks, Vol. 10, No. 3, 1999 https://doi.org/10.1109/72.761707