994 resultados para neural differentiation
Resumo:
Four types of neural networks which have previously been established for speech recognition and tested on a small, seven-speaker, 100-sentence database are applied to the TIMIT database. The networks are a recurrent network phoneme recognizer, a modified Kanerva model morph recognizer, a compositional representation phoneme-to-word recognizer, and a modified Kanerva model morph-to-word recognizer. The major result is for the recurrent net, giving a phoneme recognition accuracy of 57% from the si and sx sentences. The Kanerva morph recognizer achieves 66.2% accuracy for a small subset of the sa and sx sentences. The results for the word recognizers are incomplete.
Resumo:
Bayesian formulated neural networks are implemented using hybrid Monte Carlo method for probabilistic fault identification in cylindrical shells. Each of the 20 nominally identical cylindrical shells is divided into three substructures. Holes of (12±2) mm in diameter are introduced in each of the substructures and vibration data are measured. Modal properties and the Coordinate Modal Assurance Criterion (COMAC) are utilized to train the two modal-property-neural-networks. These COMAC are calculated by taking the natural-frequency-vector to be an additional mode. Modal energies are calculated by determining the integrals of the real and imaginary components of the frequency response functions over bandwidths of 12% of the natural frequencies. The modal energies and the Coordinate Modal Energy Assurance Criterion (COMEAC) are used to train the two frequency-response-function-neural-networks. The averages of the two sets of trained-networks (COMAC and COMEAC as well as modal properties and modal energies) form two committees of networks. The COMEAC and the COMAC are found to be better identification data than using modal properties and modal energies directly. The committee approach is observed to give lower standard deviations than the individual methods. The main advantage of the Bayesian formulation is that it gives identities of damage and their respective confidence intervals.
Resumo:
In this paper, we derive an EM algorithm for nonlinear state space models. We use it to estimate jointly the neural network weights, the model uncertainty and the noise in the data. In the E-step we apply a forwardbackward Rauch-Tung-Striebel smoother to compute the network weights. For the M-step, we derive expressions to compute the model uncertainty and the measurement noise. We find that the method is intrinsically very powerful, simple and stable.
Resumo:
This paper introduces current work in collating data from different projects using soil mix technology and establishing trends using artificial neural networks (ANNs). Variation in unconfined compressive strength as a function of selected soil mix variables (e.g., initial soil water content and binder dosage) is observed through the data compiled from completed and on-going soil mixing projects around the world. The potential and feasibility of ANNs in developing predictive models, which take into account a large number of variables, is discussed. The main objective of the work is the management and effective utilization of salient variables and the development of predictive models useful for soil mix technology design. Based on the observed success in the predictions made, this paper suggests that neural network analysis for the prediction of properties of soil mix systems is feasible. © ASCE 2011.
Resumo:
神经管闭合缺陷(NTDs)是一种严重的先天畸形疾病,在新生儿中有千分之一的发病率.神经管融合前后,多种组织参与形态发生运动.神经管一经融合,神经嵴细胞就会向背侧中线方向产生单极突出并向此方向迁移形成神经管的顶部.与此同时,神经管从腹侧开始发生辐射状切入以实现单层化.在此,我们在非洲爪蟾的移植体中机械阻断神经管的闭合以检测其细胞运动及随后的图式形成.结果显示神经管闭合缺陷的移植体不能形成单层化的神经管,并且神经嵴细胞滞留在侧面区域不能向背侧中线迁移,而对神经前体标记基因的检测显示神经管的背腹图式形成并未受到影响.以上结果表明神经管的融合对于辐射状切入和神经嵴细胞向背侧中线方向的迁移过程是必需的,而对于神经管的沿背腹轴方向的图式形成是非必需的.
Resumo:
观察了新近发现于我国云南的果蝇属暗果蝇种组( Drosophila obscura species group ) 种类 D1luguensis 、D1 dianensis 和D1limingi 的有丝分裂中期核型, 并将3 个种的核型与各自的近缘种类进行了比较。 D1luguensis 具2n = 12 条染色体, 包括3 对中央着丝粒(V 形) 染色体、2 对近端着丝粒(棒状) 染色体以及1 对微小(点状) 染色体。其中X 和Y染色体均为中央着丝粒染色体。D1 dianensis 和D1limingi 具2n = 10 条染 色体, 包括1 对大的V 形常染色体, 1 对小的V 形常染色体, 2 对J 形(亚中着丝粒型) 常染色体和1 对点状染 色体。其中X 染色体为J 形, Y染色体为短棒状。基于核型比较的结果以及D1sinobscura 亚组地理分布的资料, 结合种间系统发育关系研究结果, 认为D1 luguensis 可能保留了该亚组祖先种类的核型。D1sinobscura 的核型(2n = 12 : 2V , 1J , 2R , 1D) 可能由一个pre2“sinobscura2hubeiensis”谱系的一个分支通过臂间倒位演化而来, 而D1 hubeiensis 的核型(2n = 10 : 4V , 1D) 可能由该谱系的另一分支通过着丝粒融合(2 对近端着丝粒常染色 体的融合) 而形成。推测在D1 dianensis 和近缘欧洲种D1subsilvestris (2n = 12 : 3V , 2R , 1D) 间、D1limingi 和 东亚近缘种D1tsukubaensis (2n = 12 : 3V , 2R , 1D) 间的物种分化过程中, 可能有相似的染色体变异类型发生。
Resumo:
We compared partial sequences (402 bp) of the mitochondrial cytochrome b gene in 68 individuals of martens (Martes), weasels (Mustela) and their relatives from the Northern Hemisphere to identify the modes of geographic differentiation in each species. We then compared complete sequences (1140 bp) of the gene in 17 species of the family Mustelidae to know the spatial and temporal modes of speciation, constructing linearized trees with transversional substitutions for deeper lineage divergences and with transversions and transitions for younger lineages. Our data suggested that these lineages of Martes and Mustela differentiated in a stepwise fashion with five radiation stages from the generic divergences (stage I) to the intraspecific divergences (stage V), during the last 10 or 20 million years as the fossil evidence suggests. In the lineage of Martes, the first offshoots are of Martes flavigula, M. pennanti, and Gulo gulo (stage II), the second is M. foina (stage III), and the third are M. americana, M. martes, M. melampus, and M. zibellina (stage IV). The divergence of the lineages of Mustela is likely to have taken place concurrently with the radiations of the Martes. These divergence processes are attributable in part to the geographic allocation along the two continents, North America and Eurasia, as well as among peripheral insular domains, such as Taiwan and the Japanese Islands. In addition, the Eurasian continent itself was shown to have been involved in the species diversification in the martens and weasels.
Resumo:
To characterize the mitochondrial DNA (mtDNA) variation in Han Chinese from several provinces of China, we have sequenced the two hypervariable segments of the control region and the segment spanning nucleotide positions 10171-10659 of the coding region,
Resumo:
Drosophila lacertosa, an Oriental member of the robusta species group in the virilis-repleta radiation, has a wide distribution from northern India throughout China to the Far East. Phylogenetic analyses of mitochondrial ND2 gene sequences revealed two ge
Resumo:
During the course of evolution, the human skeletal system has evolved rapidly leading to an incredible array of phenotypic diversity, including variations in height and bone mineral density. However, the genetic basis of this phenotypic diversity and the relatively rapid tempo of evolution have remained largely undocumented. Here, we discover that skeletal genes exhibit a significantly greater level of population differentiation among humans compared with other genes in the genome. The pattern is exceptionally evident at amino acid-altering sites within these genes. Divergence is greater between Africans and both Europeans and East Asians. In contrast, relatively weak differentiation is observed between Europeans and East Asians. SNPs with higher levels of differentiation have correspondingly higher derived allele frequencies in Europeans and East Asians. Thus, it appears that positive selection has operated on skeletal genes in the non-African populations and this may have been initiated with the human colonization of Eurasia. In conclusion, we provide genetic evidence supporting the rapid evolution of the human skeletal system and the associated diversity of phenotypes.