首页 > 论文 > 中国激光 > 44卷 > 4期(pp:406003--1)

光子脉冲神经元权重器件的研制

Development of Weighting Device for Photon Spiking Neuron

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

互连的光子脉冲神经元通过权重器件联系在一起, 为了实现神经网络的大规模计算, 权重器件的实现至关重要。利用微机电系统可调光衰减器(VOA), 研制了一种可以自动调节光子脉冲神经元的权重器件。该权重器件包括VOA、光电探测器、单片机、模数转换器、数模转换器和放大器等模块, 可以根据接收的光信号快速计算查表, 可对VOA的衰减值进行实时在线调整。该权重器件效率高, 且容易实现。该权重器件配合脉冲时间依赖的可塑性(STDP)光路使用, 可以实现光子脉冲神经元的STDP学习机制。当STDP曲线窗口高度为0.2时, 对权重器件进行了测量, 实现了4种STDP学习。实验测量结果与理论计算结果一致。

Abstract

Photon spiking neurons are connected by weighting devices, so the implementation of weighting devices is critical for realizing the large-scale computation of neural networks. Based on the variable optical attenuator (VOA), a weighting device for automatically adjusting photon spiking neuron is developed. The weighting device includes VOA, photoelectric detector, single chip, analog to digital converter, digital to analog converter, amplification module, etc. The weighting device can quickly calculate and look up table, and the attenuation values of VOA can be adjusted online based on the received optical signal. The weighting device has the advantages of high efficiency and easy implementation. When we combine the optical spike-timing-dependent plasticity (STDP) circuits and the proposed weighting device, STDP learning mechanisms for photon spiking neural can be achieved. The weighting device is detected when the height of STDP curve window is 0.2, and four STDP learning curves are obtained. The experimental results are consistent with the theoretical computation results.

投稿润色
补充资料

中图分类号:TN29

DOI:10.3788/cjl201744.0406003

所属栏目:光纤光学与光通信

基金项目:国家自然科学基金(61571035, 61378061, 61401017)

收稿日期:2016-11-30

修改稿日期:2016-12-26

网络出版日期:--

作者单位    点击查看

宋晓佳:北京交通大学理学院光信息科学与技术研究所发光与光信息技术教育部重点实验室, 北京 100044
王智:北京交通大学理学院光信息科学与技术研究所发光与光信息技术教育部重点实验室, 北京 100044
李强:北京交通大学理学院光信息科学与技术研究所发光与光信息技术教育部重点实验室, 北京 100044
孙翀翚:北京交通大学理学院光信息科学与技术研究所发光与光信息技术教育部重点实验室, 北京 100044
乐燕思:北京交通大学理学院光信息科学与技术研究所发光与光信息技术教育部重点实验室, 北京 100044
崔粲:北京交通大学理学院光信息科学与技术研究所发光与光信息技术教育部重点实验室, 北京 100044
吴重庆:北京交通大学理学院光信息科学与技术研究所发光与光信息技术教育部重点实验室, 北京 100044
刘彪:北京交通大学电气工程学院, 北京 100044

联系人作者:宋晓佳(15126082@bjtu.edu.cn)

备注:宋晓佳(1993-), 女, 硕士研究生, 主要从事光纤通信和光电子器件方面的研究。

【1】McCulloch W S, Pitts W. A logical calculus of the ideas immanent in nervous activity[J]. Bulletin of Mathematical Biophysics, 1943, 5(4): 115-133.

【2】Maass W. Networks of spiking neurons: The third generation of neural network models[J]. Neural Networks, 1997, 10(9): 1659-1671.

【3】Indiveri G, Linares-Barranco Bernabé, Hamilton T J, et al. Neuromorphic silicon neuron circuits[J]. Frontiers in Neuroscience, 2011, 5: 1-23.

【4】Li Qiang, Wang Zhi, Wu Chongqing, et al. All-optical thresholder based on nonlinear optical loop mirror[J]. Chinese J Lasers, 2015, 42(7): 0705001.
李 强, 王 智, 吴重庆, 等. 基于非线性光纤环形腔镜的全光阈值器[J]. 中国激光, 2015, 42(7): 0705001.

【5】Xu Rui, Huang Daquan, Li Zhineng. The implementation of a perfect shuffle type of optoelectronic neural network system[J]. Acta Optica Sinica, 2002, 22(1): 62-66.
许 锐, 黄达诠, 李志能. 洗牌型光电混合神经网络实验系统[J]. 光学学报, 2002, 22(1): 62-66.

【6】Kravtsov K, Fok M P, Rosenbluth D, et al. Ultrafast all-optical implementation of a leaky integrate-and-fire neuron[J]. Optics Express, 2011, 19(3): 2133-2147.

【7】Fok M P, Deming H, Nahmias M, et al. Signal feature recognition based on lightwave neuromorphic signal processing[J]. Optics Letters, 2011, 36(1): 19-21.

【8】Hurtado A, Schires K, Henning I D, et al. Investigation of vertical cavity surface emitting laser dynamics for neuromorphic photonic systems[J]. Applied Physics Letters, 2012, 100(10): 103703.

【9】van Vaerenbergh T, Fiers M, Mechet P, et al. Cascadable excitability in microrings[J]. Optics Express, 2012, 20(18): 20292-20308.

【10】Alexander K, van Vaerenbergh T, Fiers M, et al. Excitability in optically injected microdisk lasers with phase controlled excitatory and inhibitory response[J]. Optics Express, 2013, 21(22): 26182-26191.

【11】Gholipour B, Bastock P, Craig C, et al. Amorphous metal-sulphide microfibers enable photonic synapses for brain-like computing[J]. Advanced Optical Materials, 2015, 3(5): 635-641.

【12】Hebb D O.The organization of behavior: A neuropsychological theory[M]. New York: Psychology Press, 2002: 1-376.

【13】Markram H, Lübke J, Frotscher M, et al. Regulation of synaptic efficacy by coincidence of postsynaptic APs and EPSPs[J]. Science, 1997, 275(5297): 213-215.

【14】Abbott L F, Nelson S B. Synaptic plasticity: Taming the beast[J]. Nature Neuroscience, 2000, 3(11s): 1178-1183.

【15】Fok M P, Tian Y, Rosenbluth D, et al. Pulse lead/lag timing detection for adaptive feedback and control based on optical spike-timing-dependent plasticity[J]. Optics Letters, 2013, 38(4): 419-421.

【16】Toole R, Fok M P. Photonic implementation of a neuronal algorithm applicable towards angle of arrival detection and localization[J]. Optics Express, 2015, 23(12): 16133-16141.

【17】Ren Q S, Zhang Y L, Wang R, et al. Optical spike-timing-dependent plasticity with weight-dependent learning window and reward modulation[J]. Optics Express, 2015, 23(19): 25247-25258.

【18】Li Q, Wang Z, LeY S, et al. Optical implementation of neural learning algorithms based on cross-gain modulation in a semiconductor optical amplifier[C]. SPIE, 2016, 10019: 100190E.

【19】Yuan Ye, Zou Yongzhuo, Bao Junfeng, et al. A novel variable optical attenuator based on micro-electromechanical systems(MEMS)[J]. Acta Optica Sinica, 2004, 24(3): 364-368.
袁 野, 邹勇卓, 鲍俊峰, 等. 一种新型微机电系统可调光衰减器[J]. 光学学报, 2004, 24(3): 364-368.

【20】Gerstner W, Kistler W M. Spiking neuron models. Single neurons, populations, plasticity[M]. Cambridge: Cambridge University Press, 2002: 1-496.

引用该论文

Song Xiaojia,Wang Zhi,Li Qiang,Sun Chonghui,Le Yansi,Cui Can,Wu Chongqing,Lin Biao. Development of Weighting Device for Photon Spiking Neuron[J]. Chinese Journal of Lasers, 2017, 44(4): 0406003

宋晓佳,王智,李强,孙翀翚,乐燕思,崔粲,吴重庆,刘彪. 光子脉冲神经元权重器件的研制[J]. 中国激光, 2017, 44(4): 0406003

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF