Unsupervised Learning of Temporal Features for Word Categorization in a Spiking Neural Network Model of the Auditory Brain

Data used for paper Higgins, I., Stringer, S. and Schnupp, J. (2016). Unsupervised Learning of Temporal Features for Word Categorization in a Spiking Neural Network Model of the Auditory Brain. PLOS ONE (in revision)<br>The data include:1. Spike rasters produced by the AN-CN-IC-CX and AN-CX models in response to four different presentations of the two words "one" and "two" pronounced by 94 different speakers 2. Distribution of delays for the AN/IC -&gt; A1 and A1 -&gt; Belt connections3. Distribution of weights for all between layer connections (AN-PL, AN-CH, AN-ON, PL-IC, CH-IC, ON-IC, IC/AN-A1, A1-Belt)本数据集用于Higgins I、Stringer S与Schnupp J（2016）的研究论文《听觉大脑脉冲神经网络模型中用于词汇分类的时序特征无监督学习》（Unsupervised Learning of Temporal Features for Word Categorization in a Spiking Neural Network Model of the Auditory Brain），该论文目前处于《PLOS ONE》的修订复审阶段。 数据集包含以下内容： 1. AN-CN-IC-CX与AN-CX模型在响应由94位不同发音者录制的单词"one"（一）与"two"（二）的四次不同呈现时，所生成的脉冲放电序列（spike rasters）； 2. AN/IC至A1以及A1至Belt连接的延迟分布； 3. 所有层间连接（AN-PL、AN-CH、AN-ON、PL-IC、CH-IC、ON-IC、IC/AN-A1、A1-Belt）的权重分布。