Neural networks have shown good results for detecting a certain pattern in a given image. In this paper, faster neural networks for pattern detection are presented. Such processors are designed based on cross correlation in the frequency domain between the input matrix and the input weights of neural networks. This approach is developed to reduce the computation steps required by these faster neural networks for the searching process. The principle of divide and conquer strategy is applied through matrix decomposition. Each matrix is divided into submatrices small in size, and then each one is tested separately by using a single faster neural processor. Furthermore, faster pattern detection is obtained by using parallel processing techniques to test the resulting submatrices at the same time, employing the same number of faster neural networks. In contrast to faster neural networks, the speed-up ratio is increased with the size of the input matrix when using faster neural networks and matrix decomposition. Moreover, the problem of local submatrix normalization in the frequency domain is solved. The effect of matrix normalization on the speed-up ratio of pattern detection is discussed. Simulation results show that local submatrix normalization through weight normalization is faster than submatrix normalization in the spatial domain. The overall speed-up ratio of the detection process is increased as the normalization of weights is done offline.
Together with the development of peritoneal dialysis (PD), appropriate animal models play an important role in the investigation of physiological, pathophysiological and clinical aspects of PD. However, there is still not an ideal experimental PD animal model. In this study, 45 Sprague-Dawley rats were divided into three groups. Grou p 1 (n=15) was receiving daily peritoneal injection through the catheter connected to the abdominal cavity, using PD solution containing 3.86 % D-glucose. Group 2 (n=15) was receiving daily peritoneal injection of 0.9 % physiological saline through a catheter. Group 3 (n=15), which was subjected to sham operation, served as controls. Our results showed that WBC counts in peritoneal effluent of Group 1 were slightly higher than those of Group 2 and control group, respectively (p<0.05). However, there was no episode of infection in any group. In addition, there was no significant difference in neutrophils fractions among these three groups. Hematoxylin-eosin and Masson’s trichrome staining demonstrated a dramatic increase in thickness of the mesothelium-to-muscle layer of peritoneum exposed to high glucose (Group 1) compared to Group 2 and controls (p<0.01). These data indicated that we established a novel rat model of PD with a modified catheter insertion method. This model is more practical, easy to operate, not too expensive and it will facilitate the investigate of long-term effects of PD., Y.-M. Peng ... [et al.]., and Obsahuje bibliografii a bibliografické odkazy
A new model for propagation of long waves including the coastal area is introduced. This model considers only the motion of the surface of the sea under the condition of preservation of mass and the sea floor is inserted into the model as an obstacle to the motion. Thus we obtain a constrained hyperbolic free-boundary problem which is then solved numerically by a minimizing method called {\em the discrete Morse semi-flow}. The results of the computation in 1D show the adequacy of the proposed model.
An original Nyquist-based frequency domain robust decentralized controller (DC) design technique for robust stability and guaranteed nominal performance is proposed, applicable for continuous-time uncertain systems described by a set of transfer function matrices. To provide nominal performance, interactions are included in individual design using one selected characteristic locus of the interaction matrix, used to reshape frequency responses of decoupled subsystems; such modified subsystems are termed "equivalent subsystems". Local controllers of equivalent subsystems independently tuned for stability and specified feasible performance constitute the decentralized controller guaranteeing specified performance of the full system. To guarantee robust stability, the M−Δ stability conditions are derived. Unlike standard robust approaches, the proposed technique considers full nominal model, thus reducing conservativeness of resulting robust stability conditions. The developed frequency domain design procedure is graphical, interactive and insightful. A case study providing a step-by-step robust DC design for the Quadruple Tank Process [K.H. Johansson: Interaction bounds in multivariable control systems. Automatica 38 (2002), 1045-1051] is included.
Community detection algorithms help us improve the management of complex networks and provide a clean sight of them. We can encounter complex networks in various fields such as social media, bioinformatics, recommendation systems, and search engines. As the definition of the community changes based on the problem considered, there is no algorithm that works universally for all kinds of data and network structures. Communities can be disjointed such that each member is in at most one community or overlapping such that every member is in at least one community. In this study, we examine the problem of finding overlapping communities in complex networks and propose a new algorithm based on the similarity of neighbors. This algorithm runs in O(mlgm) running time in the complex network containing m number of relationships. To compare our algorithm with existing ones, we select the most successful four algorithms from the Community Detection library (CDlib) by eliminating the algorithms that require prior knowledge, are unstable, and are time-consuming. We evaluate the successes of the proposed algorithm and the selected algorithms using various known metrics such as modularity, F-score, and Normalized Mutual Information. In addition, we adapt the coverage metric defined for disjoint communities to overlapping communities and also make comparisons with this metric. We also test all of the algorithms on small graphs of real communities. The experimental results show that the proposed algorithm is successful in finding overlapping communities.
This paper proposes a new quantum particle swarm optimization algorithm with local attracting (LAQPSO), which is based on quantum-inspired evolutionary algorithm (QEA) and particle swarm optimization algorithm (PSO). In the proposed LAQPSO, a novel quantum bit expression mechanism called quantum angle is employed to encode the solution onto particle, and a new local attractor is proposed to determine the rotation angle of quantum rotation gate automatically. During the process of seeking the global solution, the magnitude of rotation angle is adjusted by an important parameter called contraction coefficient, which can quantitatively determine the tradeoff between exploration ability and exploitation ability. The simulation results for different contraction coeffcients are helpful for selecting the key parameter. A set of benchmark functions are used to evaluate the performance of LAQPSO, QEA and QBPSO, and the results show that the proposed algorithm has a fast convergence rate and can effectively avoid premature convergence.
When parasites invade paired structures of their host non-randomly, the resulting asymmetry may have both pathological and ecological significance. To facilitate the detection and visualisation of asymmetric infections we have developed a free software tool, Analysis of Symmetry of Parasitic Infections (ASPI). This tool has been implemented as an R package (https://cran.r-project.org/package=aspi) and a web application (https://wayland.shinyapps.io/aspi). ASPI can detect both consistent bias towards one side, and inconsistent bias in which the left side is favoured in some hosts and the right in others. Application of ASPI is demonstrated using previously unpublished data on the distribution of metacercariae of species of Diplostomum von Nordmann, 1832 in the eyes of ruffe Gymnocephalus cernua (Linnaeus). Invasion of the lenses appeared to be random, with the proportion of metacercariae in the left and right lenses showing the pattern expected by chance. However, analysis of counts of metacercariae from the humors, choroid and retina revealed asymmetry between eyes in 38% of host fish., Matthew T. Wayland, James C. Chubb., and Obsahuje bibliografii
Embedding approaches can be used for solving non linear programs \emph{P}. The idea is to define a one-parametric problem such that for some value of the parameter the corresponding problem is equivalent to \emph{P}. A particular case is the multipliers embedding, where the solutions of the corresponding parametric problem can be interpreted as the points computed by the multipliers method on \emph{P}. However, in the known cases, either path-following methods can not be applied or the necessary conditions for its convergence are fulfilled under very restrictive hypothesis. In this paper, we present a new multipliers embedding such that the objective function and the constraints of P(t) are C3 differentiable functions. We prove that the parametric problem satisfies the \emph{JJT}-regularity generically, a necessary condition for the success of the path-following method.
In the article a new sparse low-rank matrix decomposition model is proposed based on the smoothly clipped absolute deviation (SCAD) penalty. In order to overcome the computational hurdle we generalize the alternating direction method of multipliers (ADMM) algorithm to develop an alternative algorithm to solve the model. The algorithm we designed alternatively renew the sparse matrix and low-rank matrix in terms of the closed form of SCAD penalty. Thus, the algorithm reduces the computational complexity while at the same time to keep the computational accuracy. A series of simulations have been designed to demonstrate the performances of the algorithm with comparing with the Augmented Lagrange Multiplier (ALM) algorithm. Ultimately, we apply the model to an on- board video background modeling problem. According to model the on-board video background, we can separate the video background and passenger's actions. Thus, the model can help us to identify the abnormal action of train passengers. The experiments show the background matrix we estimated is not only sparser, but the computational efficiency is also improved.