在锶(钡) 二溴对甲偶氮甲磺显色体系中,应用 Kohonen神经网络优选波长,用遗传算法优化确定BP神经网络结构和参数,得到优化结构的网络,即 KNN GA BP ANN(29 3 3 2),学习速率η=0.233 3,动量因子α=0.974 7。用优化了的神经网络解析锶、...在锶(钡) 二溴对甲偶氮甲磺显色体系中,应用 Kohonen神经网络优选波长,用遗传算法优化确定BP神经网络结构和参数,得到优化结构的网络,即 KNN GA BP ANN(29 3 3 2),学习速率η=0.233 3,动量因子α=0.974 7。用优化了的神经网络解析锶、钡配合物的混合吸收光谱,不经分离光度法同时测定锶和钡。将BP ANN 、KNN BP ANN与KNN GA BP ANN三种神经网络方法的分析结果进行比较,表明 KNN GA BP ANN最优。锶和钡的配合物的表观摩尔吸光系数分别为εSr635=6.9×104L·mol-1·cm-1,εBa634=8.0×104L·mol-1·cm-1。展开更多
The local visual motion detection mechanism used in the visual systems of primatescan only sense the motion component oriented perpendicularly to the contrast gradient of thebrightness pattern.But the visual system of...The local visual motion detection mechanism used in the visual systems of primatescan only sense the motion component oriented perpendicularly to the contrast gradient of thebrightness pattern.But the visual system of higher animals can adaptively determine the actualdirection of motion through a learning process.In this paper a multilayered feedforward neuralnetwork model for perception of visual motion is presented.This model employs W.Reichardt’selementary motion detectors array and T.Kohonen’s self-organizing feature map.We explored theself-organizing principles for perception of visual motion.The computer simulations show thatthis neural network is able to recognize the true direction of motion through an unsupervisedlearning process.In addition,the neurons with the same or similar motion direction selectivitytend to appear in“functional columns”which seem to be qualitatively similar to the corticalmotion columns observed by electrophysiological and cytohistochemical studies in certain higherareas such as MT.It proves that motion-detection by spatio-temporal coherences,mapping,co-operation,competition,and Hebb rule may be the basic principles for the self-organization ofvisual motion perception networks.展开更多
本文采用潜在语义索引(LSI)和遗传算法(GA)进行文本特征提取。在采用潜在语义索引将语义关系体现在VSM(Vector Space Model)中,通过奇异值分解(SVD,Singular Value De-composition)可以有效地降低向量空间的维数,但通过维数约简后的文...本文采用潜在语义索引(LSI)和遗传算法(GA)进行文本特征提取。在采用潜在语义索引将语义关系体现在VSM(Vector Space Model)中,通过奇异值分解(SVD,Singular Value De-composition)可以有效地降低向量空间的维数,但通过维数约简后的文本特征仍要保持在数百维左右,因此本文采用遗传算法在此基础上继续降维。实验结果表明,这两种方法结合可以极大的降低文本向量空间的维数,并能提高分类准确率。展开更多
文摘在锶(钡) 二溴对甲偶氮甲磺显色体系中,应用 Kohonen神经网络优选波长,用遗传算法优化确定BP神经网络结构和参数,得到优化结构的网络,即 KNN GA BP ANN(29 3 3 2),学习速率η=0.233 3,动量因子α=0.974 7。用优化了的神经网络解析锶、钡配合物的混合吸收光谱,不经分离光度法同时测定锶和钡。将BP ANN 、KNN BP ANN与KNN GA BP ANN三种神经网络方法的分析结果进行比较,表明 KNN GA BP ANN最优。锶和钡的配合物的表观摩尔吸光系数分别为εSr635=6.9×104L·mol-1·cm-1,εBa634=8.0×104L·mol-1·cm-1。
基金Supported in part by the National Natural Science Foundation of China National Laboratory of Pattern Recognition,Institute of Automation,Academia Sinica.
文摘The local visual motion detection mechanism used in the visual systems of primatescan only sense the motion component oriented perpendicularly to the contrast gradient of thebrightness pattern.But the visual system of higher animals can adaptively determine the actualdirection of motion through a learning process.In this paper a multilayered feedforward neuralnetwork model for perception of visual motion is presented.This model employs W.Reichardt’selementary motion detectors array and T.Kohonen’s self-organizing feature map.We explored theself-organizing principles for perception of visual motion.The computer simulations show thatthis neural network is able to recognize the true direction of motion through an unsupervisedlearning process.In addition,the neurons with the same or similar motion direction selectivitytend to appear in“functional columns”which seem to be qualitatively similar to the corticalmotion columns observed by electrophysiological and cytohistochemical studies in certain higherareas such as MT.It proves that motion-detection by spatio-temporal coherences,mapping,co-operation,competition,and Hebb rule may be the basic principles for the self-organization ofvisual motion perception networks.
文摘本文采用潜在语义索引(LSI)和遗传算法(GA)进行文本特征提取。在采用潜在语义索引将语义关系体现在VSM(Vector Space Model)中,通过奇异值分解(SVD,Singular Value De-composition)可以有效地降低向量空间的维数,但通过维数约简后的文本特征仍要保持在数百维左右,因此本文采用遗传算法在此基础上继续降维。实验结果表明,这两种方法结合可以极大的降低文本向量空间的维数,并能提高分类准确率。
