期刊
INTERNATIONAL JOURNAL OF NEURAL SYSTEMS
卷 20, 期 6, 页码 463-480出版社
WORLD SCIENTIFIC PUBL CO PTE LTD
DOI: 10.1142/S0129065710002553
关键词
STDP; supervised learning; dynamic threshold neurons; SNNs
资金
- University of Ulster, Magee Campus
This paper proposes a supervised training algorithm for Spiking Neural Networks (SNNs) which modifies the Spike Timing Dependent Plasticity (STDP) learning rule to support both local and network level training with multiple synaptic connections and axonal delays. The training algorithm applies the rule to two and three layer SNNs, and is benchmarked using the Iris and Wisconsin Breast Cancer (WBC) data sets. The effectiveness of hidden layer dynamic threshold neurons is also investigated and results are presented.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据