神經(jīng)隨機(jī)匯池網(wǎng)絡(luò)的信息傳遞研究.pdf

1、分類(lèi)號(hào)：TN911.7 密級(jí)：公開(kāi)UDC：621.39 學(xué)校代碼：11065碩士學(xué)位論文神經(jīng)隨機(jī)匯池網(wǎng)絡(luò)的信息傳遞研究 神經(jīng)隨機(jī)匯池網(wǎng)絡(luò)的信息傳遞研究王慧指 導(dǎo) 教 師 段法兵教授學(xué)科專(zhuān)業(yè)名稱(chēng) 系 統(tǒng) 理 論論文答辯日期 2017 年 5 月 24 日IIAbstractThis thesis employs the measures of the mean mutual information and thestimulus-spec

2、ific information to explore the performance of information transmissionof the stochastic pooling networks composed of saturating synaptic neurons or theIntegrate-and-Fire neurons. It is noted that the stimuli are the det

3、erministic aperiodicor the speech signals, while the internal noise is with Gamma or Gaussiandistributions. First, in saturating synaptic neural networks, it applies the aperiodicsignal as the input signal, selects Gamma

4、 noise to simulate the internal noise in nervecells. The results of the measure of the mean mutual information and thestimulus-specific information demonstrate that the noise-enhanced effect ofinformation transmission ap

5、pears in the heterogeneous stochastic pooling networkswith multi-synaptic excitatory and inhibitory pathways. Second, we also take themeasure of the mean mutual information and the stimulus-specific information toexplore

6、 the stochastic resonance effect in the Integrate-and-Fire neural stochasticpooling networks by transmitting the speech signal. The main purpose is to analyzethe effect of information transmission caused by noise intensi

7、ty and the number of theIntegrate-and-Fire neurons. Third, according to the actual data experiment, it provesthat, as the internal noise intensity increases, the neurons have a better response to theinput signal, wherein

8、 the mean mutual information can reach a maximum value in acertain range of noise intensity. It is also shown that the stimulus-specific informationcan measure the coding efficiency in each component of the input signal

9、via theinternal noise enhancement clearly. We argue that the present results are meaningfulto the information-carrying signal transmission in the neurons in future particularly.Keywords: Stochastic pooling networks; Mean