Gradient learning in spiking neural networks by dynamic perturbation of conductances
Gradient learning in spiking neural networks by dynamic perturbation of conductances
复制标题
DOI：
10.1103/physrevlett.97.048104
复制
发表时间：
2006-07-28
影响因子：
8.6
通讯作者：
Seung, H. Sebastian
中科院分区：
文献类型：
--
作者：
Fiete, Ila R.;Seung, H. Sebastian
我们提出了一种估计目标函数相对于尖峰神经网络突触权重的梯度的方法。该方法的工作原理是测量目标函数响应神经元膜电导动态扰动的波动。它与具有动态突触的基于电导的模型神经元的循环网络兼容。如果动态扰动是由外部源的随机尖峰序列驱动的一类特殊的“经验”突触生成的，则该方法可以解释为生物学上合理的突触学习规则。
We present a method of estimating the gradient of an objective function with respect to the synaptic weights of a spiking neural network. The method works by measuring the fluctuations in the objective function in response to dynamic perturbation of the membrane conductances of the neurons. It is compatible with recurrent networks of conductance-based model neurons with dynamic synapses. The method can be interpreted as a biologically plausible synaptic learning rule, if the dynamic perturbations are generated by a special class of "empiric" synapses driven by random spike trains from an external source.