Differential evolution (DE) is well known as a simple and efficient scheme for global optimization over continuous spaces. It has reportedly outperformed a few evolutionary algorithms (EAs) and other search heuristics like the particle swarm optimization (PSO) when tested over both benchmark and real-world problems. DE, however, is not completely free from the problems of slow and/or premature convergence. This paper describes a family of improved variants of the DE/target-to-best/1/bin scheme, which utilizes the concept of the neighborhood of each population member. The idea of small neighborhoods, defined over the index-graph of parameter vectors, draws inspiration from the community of the PSO algorithms. The proposed schemes balance the exploration and exploitation abilities of DE without imposing serious additional burdens in terms of function evaluations. They are shown to be statistically significantly better than or at least comparable to several existing DE variants as well as a few other significant evolutionary computing techniques over a test suite of 24 benchmark functions. The paper also investigates the applications of the new DE variants to two real-life problems concerning parameter estimation for frequency modulated sound waves and spread spectrum radar poly-phase code design.

/pdf/differential-evolution-using-a-neighborhood-based-mutation-1878quyysh.pdf

Differential Evolution Using a Neighborhood-Based Mutation Operator

In recent years the Probabilistic Neural Network (PPN) has been used in a large number of applications due to its simplicity and efficiency. PNN assigns the test data to the class with maximum likelihood compared with other classes. Likelihood of the test data to each training data is computed in the pattern layer through a kernel density estimation using a simple Bayesian rule. The kernel is usually a standard probability distribution function such as a Gaussian function. A spread parameter is used as a global parameter which determines the width of the kernel. The Bayesian rule in the pattern layer estimates the conditional probability of each class given an input vector without considering any probable local densities or heterogeneity in the training data. In this paper, an enhanced and generalized PNN (EPNN) is presented using local decision circles (LDCs) to overcome the aforementioned shortcoming and improve its robustness to noise in the data. Local decision circles enable EPNN to incorporate local information and non-homogeneity existing in the training population. The circle has a radius which limits the contribution of the local decision. In the conventional PNN the spread parameter can be optimized for maximum classification accuracy. In the proposed EPNN two parameters, the spread parameter and the radius of local decision circles, are optimized to maximize the performance of the model. Accuracy and robustness of EPNN are compared with PNN using three different benchmark classification problems, iris data, diabetic data, and breast cancer data, and five different ratios of training data to testing data: 90:10, 80:20, 70:30, 60:40, and 50:50. EPNN provided the most accurate results consistently for all ratios. Robustness of PNN and EPNN is investigated using different values of signal to noise ratio (SNR). Accuracy of EPNN is consistently higher than accuracy of PNN at different levels of SNR and for all ratios of training data to testing data.

Enhanced probabilistic neural network with local decision circles: A robust classifier

Artificial bee colony (ABC) algorithm developed by Karaboga is a nature inspired metaheuristic based on honey bee foraging behavior. It was successfully applied to continuous unconstrained optimization problems and later it was extended to constrained design problems as well. This paper introduces an upgraded artificial bee colony (UABC) algorithm for constrained optimization problems. Our UABC algorithm enhances fine-tuning characteristics of the modification rate parameter and employs modified scout bee phase of the ABC algorithm. This upgraded algorithm has been implemented and tested on standard engineering benchmark problems and the performance was compared to the performance of the latest Akay and Karaboga's ABC algorithm. Our numerical results show that the proposed UABC algorithm produces better or equal best and average solutions in less evaluations in all cases.

An upgraded artificial bee colony (ABC) algorithm for constrained optimization problems

With rapid technological innovation and strong competition in hi-tech industries such as computer and communication organizations, the upstream component price and the downstream product cost usually decline significantly with time. As a result, an effective pricing supply chain model is very important. This paper first establishes two bi-level pricing models for pricing problems with the buyer and the vendor in a supply chain designated as the leader and the follower, respectively. A particle swarm optimization (PSO) based algorithm is developed to solve problems defined by these bi-level pricing models. Experiments illustrate that this PSO based algorithm can achieve a profit increase for buyers or vendors if they are treated as the leaders under some situations, compared with the existing methods.

/pdf/particle-swarm-optimization-for-bi-level-pricing-problems-in-3o110wmhn7.pdf

Particle swarm optimization for bi-level pricing problems in supply chains

Example-based learning particle swarm optimization for continuous optimization

An improved particle swarm optimization (IPSO) was proposed in this paper to solve the problem that the linearly decreasing inertia weight (LDIW) of particle swarm optimization algorithm cannot adapt to the complex and nonlinear optimization process The strategy of nonlinear decreasing inertia weight based on the concave function was used in this algorithm The aggregation degree factor of the swarm was introduced in this new algorithm And in each iteration process, the weight is changed dynamically based on the current aggregation degree factor and the iteration times, which provides the algorithm with dynamic adaptability The experiments on the three classical functions show that the convergence speed of IPSO is significantly superior to LDIWPSO, and the convergence accuracy is increased

An Improved Particle Swarm Optimization Algorithm

In emergency, the crowd evacuation from public buildings is the most important issue to save human lives. Panic generation and spread normally can lead to the unstable state -stampede during the crowd motion. The stability of crowd evacuation is a complex problem being researched for decades. This paper introduces self-organized criticality(SOC) theory to build the mapping model from a collective crowd into a sand pile with SOC. Therefore, the complex problem of stability analysis for crowd evacuation is converted into sandpiper stability analysis in a relatively simpler way.

Crowd Evacuation Stability Based on Self-Organized Criticality Theory

Nowadays the deteriorating global climate alters us to lower the energy consumption and reduce the emission simultaneously In the aspect of economy, energy-saving and emission-reduction are feasible methods to lower the product cost, and to obtain more profits for a factory or an enterprise There are increasing energy-saving and emission-reduction technologies designed by governments, device manufacturers, software companies, research institutes Thereby a real problem arising is how to select the appropriate solutions for a factory or an enterprise according to their real production characteristics and different duration plan for their future development Cloud service, as a novel service mode, supplies a new solution for the problem of selection and realization about the possible technologies to save energy consumption and reduce the emission from the real production line Therefore this paper proposes a technical solution of service cloud building for energy-emission and emission-reduction in manufacturing industry Then main function modules are designed, and this paper explains the key-technologies to build this service cloud respectively Finally some useful conclusions are drawn for the real application of service cloud in the engineering respectively

A Technical Solution of Service Cloud Building for Energy-Saving and Emission-Reduction in Manufacture Industry

According to the necessity for researching the optimization of enterprise energy-consuming based on model, the identification method for energy-consuming link in enterprise production process was researched. In view of the existing problems of Least Squares Support Vector Machine (LS-SVM), a modeling method based on Liu-transformation and LS-SVM was proposed. Liu-transformation was applied to intelligent data analysis for extracting typical characteristics from the training samples, and then the data were trained to construct the energy-consuming model based on on-line LS-SVM algorithm. The simulation result showed that the presented modeling method has the advantages of shorter computing time, robust and better generalization ability.

Industrial Process Modeling Based on Online Learning Algorithm for Regression Least Squares Support Vector Machine

Recently, the intrinsic magnetic topological insulator MnBi2Te4 has attracted enormous research interest due to the great success in realizing exotic topological quantum states, such as the quantum anomalous Hall effect (QAHE), axion insulator state, high-Chern-number and high-temperature Chern insulator states. One key issue in this field is to effectively manipulate these states and control topological phase transitions. Here, by systematic angle-dependent transport measurements, we reveal a magnetization-tuned topological quantum phase transition from Chern insulator to magnetic insulator with gapped Dirac surface states in MnBi2Te4 devices. Specifically, as the magnetic field is tilted away from the out-of-plane direction by around 40-60 degrees, the Hall resistance deviates from the quantization value and a colossal, anisotropic magnetoresistance is detected. The theoretical analyses based on modified Landauer-Buttiker formalism show that the field-tilt-driven switching from ferromagnetic state to canted antiferromagnetic state induces a topological quantum phase transition from Chern insulator to magnetic insulator with gapped Dirac surface states in MnBi2Te4 devices. Our work provides an efficient means for modulating topological quantum states and topological quantum phase transitions. 

/pdf/magnetization-tuned-topological-quantum-phase-transition-in-1lu0by8m.pdf

Magnetization-tuned topological quantum phase transition in 
<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mi>MnBi</mml:mi><mml:mn>2</mml:mn></mml:msub><mml:msub><mml:mi>Te</mml:mi><mml:mn>4</mml:mn></mml:msub></mml:mrow></mml:math>
 devices

Jian Wang

Papers

An Improved Particle Swarm Optimization Algorithm

Crowd Evacuation Stability Based on Self-Organized Criticality Theory

A Technical Solution of Service Cloud Building for Energy-Saving and Emission-Reduction in Manufacture Industry

Industrial Process Modeling Based on Online Learning Algorithm for Regression Least Squares Support Vector Machine

Magnetization-tuned topological quantum phase transition in <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mi>MnBi</mml:mi><mml:mn>2</mml:mn></mml:msub><mml:msub><mml:mi>Te</mml:mi><mml:mn>4</mml:mn></mml:msub></mml:mrow></mml:math> devices