Spike based normalizing Hebbian learning in an analog VLSI artificial neuron

被引:8
|
作者
Häfliger, P [1 ]
Mahowald, M [1 ]
机构
[1] ETH Zurich, UNIZ, Inst Neuroinformat, CH-8006 Zurich, Switzerland
关键词
temporal correlation; spike based learning; Hebbian learning rule; neuromorphic hardware; Riccati equation;
D O I
10.1023/A:1008351103449
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Recent neurophysiological results indicate that changes in synaptic efficacy are dependent on cooccurrence of a pre and a postsynaptic spike at the synapse [5,8]. There are only a few models of parts of the nervous system that use temporal correlation of single spikes in learning [1]. Ln most models of artificial neural networks neurons communicate by analog signals representing frequencies, and their learning rules are also defined on these continuous signals. Timing of single spikes is not used, nor is it represented. This simplification has proven useful in many applications and it makes simulations in software simpler and faster. Spiking systems have been avoided because they are computationally more difficult. However, by implementing spiking and learning artificial neurons in analog VLSI it is possible to examine the behavior of these more detailed models in real time. This is why ourselves and others [4] have started to use silicon models of spiking learning neurons. We have formulated one possible mechanism of weight normalization: a Hebbian learning rule that makes use of temporal correlations between single spikes. We have implemented it on a CMOS chip and demonstrate its normalizing behavior.
引用
收藏
页码:133 / 139
页数:7
相关论文
共 50 条
  • [31] Spike-based learning with a generalized integrate and fire silicon neuron
    Indiveri, Giacomo
    Stefanini, Fabio
    Chicca, Elisabetta
    2010 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS, 2010, : 1951 - 1954
  • [32] Active spike responses of analog electrical neuron: Theory and experiments
    Binczak, S.
    Nguetcho, A. S. Tchakoutio
    Jacquir, S.
    Bilbault, J. M.
    Kazantsev, V. B.
    2010 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS, 2010, : 2550 - 2553
  • [33] A New Compact Analog VLSI Model for Spike Timing Dependent Plasticity
    Azghadi, Mostafa Rahimi
    Al-Sarawi, Said
    Iannella, Nicolangelo
    Abbott, Derek
    2013 IFIP/IEEE 21ST INTERNATIONAL CONFERENCE ON VERY LARGE SCALE INTEGRATION (VLSI-SOC), 2013, : 7 - 12
  • [34] A low power density hebbian eigenfilter VLSI for neuro-physiological sensor spike train analysis
    Peng, Liang
    Li, Xiangyu
    Yu, Bo
    Mak, Terrence
    Sun, Yihe
    CHINA SEMICONDUCTOR TECHNOLOGY INTERNATIONAL CONFERENCE 2012 (CSTIC 2012), 2012, 44 (01): : 1407 - 1412
  • [35] Temporally asymmetric Hebbian learning, spike timing and neuronal response variability
    Abbott, LF
    Song, S
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 11, 1999, 11 : 69 - 75
  • [36] Spike-timing-dependent Hebbian plasticity as temporal difference learning
    Rao, RPN
    Sejnowski, TJ
    NEURAL COMPUTATION, 2001, 13 (10) : 2221 - 2237
  • [37] Analysis and application of an artificial neuron analog circuit
    Tianjin Univ, Tianjin, China
    Tien Tzu Hsueh Pao, 11 (109-111, 108):
  • [38] Artificial neural network in analog VLSI-technology
    Geske, G
    Stüpmann, R
    Rode, S
    BEC 2002: Proceedings of the 8th Biennial Baltic Electronic Conference, 2002, : 115 - 118
  • [39] Stable Hebbian learning from spike timing-dependent plasticity
    van Rossum, MCW
    Bi, GQ
    Turrigiano, GG
    JOURNAL OF NEUROSCIENCE, 2000, 20 (23): : 8812 - 8821
  • [40] Spiking Analog VLSI Neuron Assemblies as Constraint Satisfaction Problem Solvers
    Binas, Jonathan
    Indiveri, Giacomo
    Pfeiffer, Michael
    2016 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS (ISCAS), 2016, : 2094 - 2097