基于復合型神經(jīng)網(wǎng)絡(luò)的非線性ICA及其在SCP少次提取中應(yīng)用研究
Nonlinear ICA Based on a Combined Neural Network and Its Application to Single-Trial Extraction of SCP
-
摘要: 該文提出一種基于MISEP和NLFA方法的復合無監(jiān)督多層感知神經(jīng)網(wǎng)絡(luò)模型解決非線性獨立分量分析(ICA)的解混問題,并對MISEP神經(jīng)網(wǎng)絡(luò)中用到的兩種Sigmoid函數(shù)及新引入的徑向基函數(shù)(RBF)作了信號分離性能的對比分析。實驗結(jié)果表明,本文方法可以更好地從非線性混合信號中復現(xiàn)源信號,穩(wěn)定性高,同時應(yīng)用于慢皮層電位(SCP)的少次提取,經(jīng)與相干平均法比較,波形的整體提取效果明顯。
-
關(guān)鍵詞:
- 非線性ICA; 互信息; 全體學習; 慢皮層電位
Abstract: A combined, unsupervised, multilayer perceptron neural network model based on MISEP and NLFA is presented to resolve the separation problem of nonlinear ICA, the separation performances of signals are compared between two sigmoid functions used in the latent layers of MISEP and introduced RBF. Experimental results show this algorithm can recover sources from nonlinear mixtures better and has good stabilization, it is also applied to single-trial extraction of SCP, the whole effect is evident compared with the averaged method. -