Deterministic and probabilistic occupancy detection with a novel heuristic optimization and Back-Propagation (BP) based algorithm

In this paper, a novel hybrid approach for deterministic and probabilistic occupancy detection is proposed with a novel heuristic optimization and Back-Propagation (BP) based algorithms. Generally, PB based neural network (BPNN) suffers with the optimal value of weight, bias, trapping problem in local minima and sluggish convergence rate. In this paper, the GSA (Gravitational Search Algorithm) is implemented as a new training technique for BPNN is order to enhance the performance of the BPNN algorithm by decreasing the problem of trapping in local minima, enhance the convergence rate and optimize the weight and bias value to reduce the overall error. The experimental results of BPNN with and without GSA are demonstrated and presented for fair comparison and adoptability. The demonstrated results show that BPNNGSA has outperformance for training and testing phase in form of enhancement of processing speed, convergence rate and avoiding the trapping problem of standard BPNN. The whole study is analyzed and demonstrated by using R language open access platform. The proposed approach is validated with different hidden-layer neurons for both experimental studies based on BPNN and BPNNGSA.

TRAPPING PROBLEM OF THE WEIGHTED SCALE-FREE TRIANGULATION NETWORKS FOR BIASED WALKS

In this paper, we study the trapping problem in the weighted scale-free triangulation networks with the growth factor [Formula: see text] and the weight factor [Formula: see text]. We propose two biased walks, one is standard weight-dependent walk only including the nearest-neighbor jumps, the other is mixed weight-dependent walk including both the nearest-neighbor and the next-nearest-neighbor jumps. For the weighted scale-free triangulation networks, we derive the exact analytic formulas of the average trapping time (ATT), the average of node-to-trap mean first-passage time over the whole networks, which measures the efficiency of the trapping process. The obtained results display that for two biased walks, in the large network, the ATT grows as a power-law function of the network size [Formula: see text] with the exponent, represented by [Formula: see text] when [Formula: see text]. Especially when the case of [Formula: see text] and [Formula: see text], the ATT grows linear with the network size [Formula: see text]. That is the smaller the value of [Formula: see text], the more efficient the trapping process is. Furthermore, comparing the standard weight-dependent walk with mixed weight-dependent walk, the obtained results show that although the next-nearest-neighbor jumps have no main effect on the trapping process, they can modify the coefficient of the dominant term for the ATT. The smaller the value of probability parameter [Formula: see text], the more efficient the trapping process for the mixed weight-dependent walk is.

Bayesian estimation of ordinary differential equation models when the likelihood has multiple local modes

Ordinary differential equations (ODEs) are popularly used to model complex dynamic systems by scientists; however, the parameters in ODE models are often unknown and have to be inferred from noisy measurements of the dynamic system. One conventional method is to maximize the likelihood function, but the likelihood function often has many local modes due to the complexity of ODEs, which makes the optimizing algorithm be vulnerable to trap in local modes. In this paper, we solve the global optimization issue of ODE parameters with the help of the Stochastic Approximation Monte Carlo (SAMC) algorithm which is shown to be self-adjusted and escape efficiently from the “local-trapping” problem. Our simulation studies indicate that the SAMC method is a powerful tool to estimate ODE parameters globally. The efficiency of SAMC method is demonstrated by estimating a predator-prey ODEs model from real experimental data.

Air trapping problem during infiltration on the large areas

The process of flow modeling in unsaturated porous medium is often found in many fields of sciences: geology, fluid mechanics, thermodynamics, microbiology or chemistry. Problem is relatively complicated due to complexity of the system which contains three phases: water, air and soil skeleton. The flow of water in such a medium can be described using two-phase (2PH) flow formulation, which accounts the inflow of air and water phases, or with simplified model known as Richards (RE) equation where only water flow is taken into account. In many well known programs available in the market (like SeepW, STOMP) the primary interest is only the water flow and the flow of air is omitted. As a result Richard equation in used more often. It’s main assumption is that pore air is continuous and has connection with atmospheric air which is equivalent to infinite mobility of the air phase during all simulation. This paper presents a brief review of the influence of the air phase in soil on water flow and pore pressure generation, with focus on applications related to infiltration process occurring in the large areas. An irrigation effect of rice fields with shallow water table has been investigated. To assess the impact of the gas phase various lengths of the infiltration zone have been considered. Numerical simulations are carried out to investigate the differences between the Richards equation and the two-phase flow model, using an in-house code based on the finite volume method.

Determining average path length and average trapping time on generalized dual dendrimer

Dendrimer has wide number of important applications in various fields. In some cases during transport or diffusion process, it transforms into its dual structure named Husimi cactus. In this paper, we study the structure properties and trapping problem on a family of generalized dual dendrimer with arbitrary coordination numbers. We first calculate exactly the average path length (APL) of the networks. The APL increases logarithmically with the network size, indicating that the networks exhibit a small-world effect. Then we determine the average trapping time (ATT) of the trapping process in two cases, i.e., the trap placed on a central node and the trap is uniformly distributed in all the nodes of the network. In both case, we obtain explicit solutions of ATT and show how they vary with the networks size. Besides, we also discuss the influence of the coordination number on trapping efficiency.