Implementation of Ultra Micro Boltzmann Machine Neuron

超微型玻尔兹曼机神经元的实现

• 批准号: 09650375
• 负责人: AKAZAWA Masamichi
• 金额: $ 2.5万
• 依托单位: HOKKAIDO UNIVERSITY
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (C)
• 财政年份: 1997
• 资助国家: 日本
• 起止时间: 1997 至 1998
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-09650375/
• 关键词: Single-Electron Tunneling; Neural Network; Boltzmann Machine; Hopfield Network; Neuron

We have investigated ways of using a single-electron-tunneling (SET) circuit to solve problems in neural networks, and have obtained the following results :1) A simple circuit for a Boltzmann machine neuron : The circuit for the Boltzmann machine neuron that has a stochastic response is usually complicated. We found, however, that by using a single-electron-tunneling circuit, we can construct a compact neuron circuit in which the Coulomb blockade condition is adjusted appropriately, It can produce an output of a random 1-0 bit stream with the probability for an output of 1 controlled by an input signal. A network consisting of more than two neurons can be constructed. So called 'annealing' operation of the network is available by controlling the bias voltage.2) A way to eliminate the local-minimum problem : If we utilize the cotunneling phenomenon found in single-electron circuits, we can obtain a neural network without the local-minimum problem. In this network, two or more threshold elements can change their outputs simultaneously in a coherent combination. Therefore a state transition with a large Hamming distance can occur and the local-minimum difficulty can disappear. As a result, the global-minimum energy state can always be achieved. Use of this property makes possible the development of novel computation devices that solve combinatorial problems without hindrance from the local-minimum difficulty.

我们研究了利用单电子隧穿（SET）电路解决神经网络问题的方法，并得到了以下结果：1）玻尔兹曼机器神经元的简单电路：具有随机响应的玻尔兹曼机器神经元的电路通常是复杂的。然而，我们发现，通过使用单电子隧穿电路，我们可以构建一个紧凑的神经元电路，其中库仑阻塞条件进行适当的调整，它可以产生一个随机的1-0比特流的输出与概率为1的输出由输入信号控制。可以构造由两个以上神经元组成的网络。通过控制偏置电压，可以实现网络的退火操作。2）消除局部最小值问题的方法：利用单电子电路中的共隧穿现象，可以得到不存在局部最小值问题的神经网络。在该网络中，两个或多个阈值元件可以在相干组合中同时改变它们的输出。因此，可以发生具有大汉明距离的状态转移，并且局部最小困难可以消失。因此，总是可以实现全局最小能量状态。利用这一特性，可以开发新的计算设备，解决组合问题，而不受局部最小困难的阻碍。

项目成果

Akazawa M.: "Quantum Hopfield Network Using Single-Electron Circuits" Extended Abstracts of the 1997 International Conference on Solid State Devices and Materials (Hamamatsu, September 16-19,1997). 306-307 (1997)

Akazawa M.：“使用单电子电路的量子 Hopfield 网络”1997 年国际固态器件和材料会议的扩展摘要（滨松，1997 年 9 月 16-19 日）。

Akazawa, M., and Amemiya Y.: "Boltzmann machine neuron circuit using single-electron tunneling" Appl.Phys.Lett.70 (5). 670-672 (1997)

Akazawa, M. 和 Amemiya Y.：“使用单电子隧道的玻尔兹曼机神经元电路”Appl.Phys.Lett.70 (5)。

Akazawa,M.: "“Quantum Hopfield Network Using Single-Electron Circuits-A Novel Hopfield Network Free from the Local -Minimum Difficulty"" accepted for publication in Analog Integrated Circuits and Signal Processing,1999.(1999)

Akazawa, M.：““Quantum Hopfield Network using Single-Electron Circuits-A Novel Hopfield Network Free from the Local -Minimum Difficulty””接受发表于《模拟集成电路和信号处理》，1999 年。(1999)

Akazawa, M.: ""Annealing method for operating quantum-cellular-automaton systems"" J.Appl.Phys.82(10). 5176-5184 (1997)

Akazawa, M.：“用于操作量子细胞自动机系统的退火方法”J.Appl.Phys.82(10)。

Akazawa, M.: ""Boltzmann machine neuron circuit using single-electron tunneling"" Appl.Phys.Lett. 70(5). 670-672 (1997)

Akazawa, M.：“使用单电子隧道效应的玻尔兹曼机神经元电路”Appl.Phys.Lett。