Constraints in Particle Swarm Optimization of Hidden Markov Models

Size: px
Start display at page:

Download "Constraints in Particle Swarm Optimization of Hidden Markov Models"

Transcription

1 Constraints in Particle Swarm Optimization of Hidden Markov Models Martin Macaš, Daniel Novák, and Lenka Lhotská Czech Technical University, Faculty of Electrical Engineering, Dep. of Cybernetics, Prague, Czech Republic Abstract. This paper presents new application of Particle Swarm Optimization (PSO) algorithm for training Hidden Markov Models (HMMs). The problem of finding an optimal set of model parameters is numerical optimization problem constrained by stochastic character of HMM parameters. Constraint handling is carried out using three different ways and the results are compared to Baum-Welch algorithm (BW), commonly used for HMM training. The global searching PSO method is much less sensitive to local extremes and finds better solutions than the local BW algorithm, which often converges to local optima. The advantage of PSO approach was markedly evident, when longer training sequence was used. 1 Introduction Hidden Markov Model (HMM) is statistical modelling technique which enables to characterize real-world stochastic signals in terms of signal models. There are many real world problems which have been solved by HMMs. The examples are speech recognition [1], biomedical signal processing, bioinformatics, robotics, handwriting recognition or economy. There are two main problems connected with HMMs - the choice of model topology (e.g. number of hidden states and type of connections) and search for the model parameters. This problem is mostly solved using Baum-Welch algorithm (BW). The main drawback of the method is frequent convergence to local optima resulting from local character of searching. One solution of this problem is to use some global searching methods [2]. In this paper, the model parameters are searched while the topology is given in advance for all experiments. The model parameters are searched in terms of likelihood maximization. The problem of reaching the global maxima is solved using nature inspired Particle Swarm Optimization (PSO) method, which is described below. However, the problem of searching some parameters of HMM is made difficult by constraints arising from the probabilistic character of this type of parameters. Furthermore, some other parameters representing covariance are searched and therefore a positivity constraint arises from this fact. This implies a need for some constraint handling mechanism to be used in PSO. E. Corchado et al. (Eds.): IDEAL 2006, LNCS 4224, pp , c Springer-Verlag Berlin Heidelberg 2006

2 1400 M. Macaš, D. Novák, and L. Lhotská 2 Methods 2.1 Hidden Markov Models An HMM is a stochastic finite state automata defined by the following parameters: λ =(A, p, B), where A = {a ij } is a matrix of state transition probabilities a ij representing probability of transition from state i to state j, p is the initial state probability and B is the emission probability density function of each state. This paper considers continuous HMMs with B represented by a finite mixture of Gaussian distributions. For simplicity, only one Gaussian distribution was used for all experiments and therefore the B is set of N Gaussian distributions G i (m i,u i ), where the m i and U i are mean and variance of the distribution for ith state respectively. Consider an observation sequence O = O 1,...,O T. There are three problems connected with HMM. The first one is to compute P (O λ), the probability of observing the sequence, given the model (model evaluation). The second problem is to find a corresponding state sequence Q = Q 1,...,Q T, given the observation sequence O and model parameters λ. The third problem is crucial and by far the most difficult one. How to find the model parameters λ maximizing P (O λ) (fitting the training sequence). This problem of model training consists of adjusting the model parameters λ to maximize likelihood - the probability of the observation sequence given the model. There is no known way to analytically solve this problem. Usually the alternative solutions are iterative procedures like Baum-Welch algorithm [3]. Baum-Welch algorithm [1] is a standard solution for solving the training problem based on maximization of data likelihood. However, this hill-climbing algorithm leads to local maxima only, and in most problems of interest, the optimization surface is very complex and has many local optima. The BW algorithm uses a set of re-estimation formulas and guarantees that the likelihood of re-estimated model will be equal to or greater than the likelihood of model in previous iteration. 2.2 Particle Swarm Optimization The PSO method is one of optimization method developed for finding a global optima of some nonlinear function [4]. It has been inspired by social behavior of birds and fish. The method applies the approach of problem solving in groups. Each solution consists of set of parameters and represents a point in multidimensional space. The solution is called particle and the group of particles (population) is called swarm. Each particle i is represented as a D-dimensional position vector x i (t) and has a corresponding instantaneous velocity vector v i (t). Furthermore, it remembers its individual best value of fitness function and position p i which has resulted in that value. During each iteration t, the velocity update rule (1) is applied on each particle in the swarm. The p g is the best position of the entire swarm and represents the social knowledge. v i (t) =w v i (t 1) + ϕ 1 R 1 ( p i x i (t 1)) + ϕ 2 R 2 ( p g x i (t 1)) (1)

3 Constraints in Particle Swarm Optimization of HMMs 1401 The parameter w is called inertia weight and during all iterations decreases linearly from w start to w end.thesymbolsr 1 and R 2 represent the diagonal matrices with random diagonal elements drawn from a uniform distribution between 0 and 1. The parameters ϕ 1 and ϕ 2 are scalar constants that weight influence of particles own experience and the social knowledge. Next, the position update rule (2) is applied: x i (t) = x i (t 1) + v i (t). (2) If any component of v i is less than V max or greater than +V max,thecorresponding value is replaced by V max or +V max, respectively. The V max is maximum velocity parameter. The update formulas (1) and (2) are applied during each iteration and the pi and p g values are updated simultaneously. The algorithm stops if maximum number of iterations is achieved. 2.3 Particle Structure and Evaluation The quality of particle (candidate solution) is evaluated using a fitness function. Each particle is built of A, p,m,u parameters reconfigured to one dimensional vector. Two sorts of structure of particles (encoding) were used here (see the Table 1) - the Type A encoding is used when the particle corresponds to all parametersλ, the type B encoding does not include last component of probabilistic parameters, because it can be computed from stochastic constraints described in the next section. Particles of both types are equivalent to parameters λ of HMM. The Type A representation was used for repair based method and the Type B encoding was used for penalty based method. Table 1. Two types of representation of a particle Type A A{N N} p{n} m{n} U{N} a a 1N... a N1... a NN p 1... p N m 1... m N U 1... U N Type B A{N (N 1)} p{n 1} m{n} U{N} a a 1N 1... a N1... a NN 1 p 1... p N 1 m 1... m N U 1... U N In this paper, the maximization of likelihood is performed as minimization of the value of function described by Equation Constraint Handling eval F (λ) = ln P (O λ) (3) An important aspect of the Baum-Welch algorithm is that the stochastic constraints of the HMM parameters are automatically satisfied at each iteration.

4 1402 M. Macaš, D. Novák, and L. Lhotská Generally, this is not true for all training algorithms. The problem of training must be therefore understood as constraint optimization task and could be described as following: Find λ = {A, p, B}, which maximizes P (O λ), subject to a ik 0 (4) N a ij =1 (5) j=1 p j 0 (6) N p j =1 (7) j=1 U i > 0, (8) where i =1...N, k =1...N and N is number of hidden states. However, the PSO was originally developed for unconstrained optimization and hence the algorithm described in previous section must be modified to be able to handle with constraints [5]. In this paper, the following constraint handling methods are compared: Methods based on repair algorithm (PSOr, PSOrf). The unfeasible solutions are repaired by moving them into feasible space. The unfeasible solutions can be replaced by its repaired version or the repair procedure can be used for fitness evaluation only. The two following repair methods are used in this paper. First approach, reffered to as PSOr, replaces unfeasible particles by their repaired version. The second one, referred to as PSOrf, uses modification of fitness function, which evaluates the repaired version of unfeasible particle, however, the unfeasible particles are not replaced. The repair procedure consists of normalization operations. First, all negative elements of state transition matrix A are assigned to be zero. Next, each element a ij of the matrix is transformed to a ij / N j=1 a ij which ensures that the constraint (5) is not violated. The same procedure is applied on initial probabilities p. Finally, a violation of positivity constraint (8) (for multivariate observations, it is constraint of positive-definity) is repaired by setting all C i which are smaller than a constant c (e.g. c =10 5 )toc. Method based on penalty function (PSOp). The unfeasible solutions are penalized using eval U (λ) =eval F (λ)+penalty(λ). However, the likelihood can not be computed for model parameters violating the stochastic constraints. Therefore, a constant K couldbeusedinplaceofeval F (λ). For instance, the suitable value for K could be the fitness function of the worst particle in the initial swarm. The following form of fitness evaluation was used: eval U (λ) =K + ξ[( (c U i ) 2. (9) a ij<0 a 2 ij + p i<0 p 2 i )] + U i<c

5 Constraints in Particle Swarm Optimization of HMMs 1403 The constant c in Equation 9 has similar meaning as in the previous section and was set to c =10 5. The penalty term computes the degree of constraint violation. The constant ξ is weight which scales an influences of stochastic constraints (4, 6) and the positivity constraint (8). The suitable value of ξ is determined by the observation sequence and was set to ξ = 10 for all experiments. The implementation of penalty function uses different conception of feasibility. The parameters of HMM are encoded into particle using the Type B encoding. Thus, the last column of transition matrix A and last component of vector of initial state probabilities are removed from particle. In the beginning of each evaluation, these parameters must be computed using Equations 5 and 7. For instance, the p N parameter is p N =1 N 1 i=1 p i. Thus, the feasibility is equivalent to satisfaction of the Equations 4, 6 and 8. In the case of violating these three inequality constraints, the particle is evaluated using Equation 9. In the case, that the particle represents feasible solution, it is evaluated using log-likelihood function (3). In experimental results, the penalty method is referred to as PSOp. 3 Results The approaches described above were tested using three observation sequences generated by three predefined artificial HMMs (Model 1-3 ). The number of hidden states was N = 4 because of two reasons. First, authors of this work will utilize the results in application of HMM to classification of 4 substages of non- REM sleep from sleep EEG record. Secondly, the PSO method fits to optimize smaller HMMs as it will be discussed at the end of this section. The initial state probabilities were set uniform (to 0.25) for all three models (p 1 = p 2 = p 3 = (0.25, 0.25, 0.25, 0.25)). For the models 1 and 2, the transition probabilities were set also uniformly to On the other hand, the diagonal elements of matrix A 3 (probabilities of staying in the previous state) were 0.85 and the others (probabilities of transition to different state) were set to The Gaussian distributions representing the emission probabilities are described in Equation 10. G( 2, 0.5) G(0, 0.2) G(0, 0.2) B 1 = G(0, 0.5) G(2, 0.5),B 2 = G(0, 2.0) G( 4, 0.5),B 3 = G(0, 2.0) G( 4, 0.5) (10) G(4, 0.5) G(4, 0.5) G(4, 0.5) The BW algorithm and the three constraint handling PSO methods were compared. The comparison criterion was the final log-likelihood function described by Equation 3 averaged over 30 runs of experiment. First, a training observation sequence of length T was generated by the model (T was 100 and 200). Further, the optimization algorithm was launched 30 times searching for the model which best fits the training sequence. For all PSO experiments, the algorithm parameters were set as following: the weight parameters were ϕ 1 = ϕ 2 = 2, number of iterations was 500, swarm size was 100. The inertia weight w has decreased linearly from w =1tow =0.2

6 1404 M. Macaš, D. Novák, and L. Lhotská during 350 iterations and then it remained constant for the rest of optimization process. The maximum velocity parameter was set variously for different dimensions. Concretely, the change of probabilistic parameters representing the transition matrix and initial state probability was limited by V max =0.2, the rest of velocity dimensions was limited by V max =2.ThevalueofV max is problem dependent and was selected experimentally. Table 2. Experimental results: means, standard deviations of ln P (O λ) obtained from 30 runs of experiments and the results of t-test (α =0.05) of difference between BW results and results for particular PSO modifications. The 1 behind the slash means that the PSO method found significantly better solutions than the BW algorithm. T = 100 Method Model 1 Model 2 Model 3 BW (4.71)/ (5.66)/ (27.13)/- PSOr (2.67)/ (5.46)/ (1.54)/1 PSOrf (3.50)/ (4.69)/ (2.03)/1 PSOp (2.04)/ (4.87)/ (2.46)/1 T=200 Model 1 Model 2 Model 3 BW (3.46)/ (22.52)/ (30.54)/- PSOr (3.44)/ (3.02)/ (2.64)/1 PSOrf (5.22)/ (6.40)/ (1.66)/1 PSOp (1.26)/ (0.72)/ (1.65)/1 The final values of ln P (O λ) obtained for the three models fitting the sequences of two different lengths (T = 100 and T = 200) are depicted in Table 2. The Figure 1 shows the training curves averaged over 30runs. The three PSO variants can be compared from this figure. The difference between BW and PSO algorithms can be seen from the Table 2. The standard deviation of final fitness value represents the ability of method to reach the minimum independently on initialization. In addition, the test of significance was performed using paired t-test (α =0.05) of the hypothesis that two matched samples X, Y come from distributions with equal means. However, the test makes an assumption that the distribution of X Y comes from a normal distribution. This assumption was not satisfied for some samples (e.g. comparison for Model 3) and hence the testing was just additional comparison tool. For all experiments, the means of final fitness obtained for PSO methods are smaller. For the short training sequence (T = 100), only the Model 3 was identified significantly better by all PSO approach. On the other hand, if the longer training sequence is used for training, the PSO algorithms reached much smaller values of the function eval F than the BW algorithm.

7 Constraints in Particle Swarm Optimization of HMMs 1405 (a) (b) Model 1 Model Model 2 Model Model 3 PSOp PSOr PSOrf Model 3 PSOp PSOr PSOrf 420 Fig. 1. The training curves averaged over 30 runs of experiment. An observation sequence with length (a) T = 100 and (b) T = 200 samples was used for training the HMM. Further, the different PSO constraint handling methods may be compared using the Table 2 and the Figure 1. The figure depicts the optimization process using the dependence of the best fitness in the swarm on time. The curves are averaged over 30 runs of experiment. The figures show strong similarity between the two repair methods (PSOr, PSOrf). At the beginning of optimization, the curve decreases faster for the PSOr and PSOrf methods, however, for PSOp, the rapidity of fall increases during first iterations. Furthermore, it even outruns the repair methods and reaches better value. Apparently, the concave-like shape of first part of training is caused by increasing number of feasible particles in the swarm. Finally, the penalty method PSOp seems least sensitive to initialization, because the minimum is reached with the smallest standard deviation and the difference of mean of results from the BW results was statistically significant. Finally, some temporal aspects of the methods must be mentioned. The main disadvantage of all population based techniques are time requirements. During each iteration of PSO using a swarm of 100 particles, the calculation of fitness function must be repeated 100 time. During the whole optimization process with 500 iterations, there are fitness evaluations. The fitness evaluation depends on the order of HMM and on length of the observation sequence. Therefore, it can be concluded, that PSO should be used for HMM training in the case of reasonable model order and not very long training sequence. The detailed experiments with time requirements were not performed for this paper.

8 1406 M. Macaš, D. Novák, and L. Lhotská 4 Conclusions The Particle Swarm Optimization seems useful and suitable method for Hidden Markov Models training. In comparison with conventional method - Baum-Welch algorithm, the PSO is much more capable to overcome the local optima of likelihood, whereas the local BW algorithm sometimes gets stuck in local extreme. In the future, the other methods will be tested. Namely, the preserving feasibility methods using Linear PSO [6] or pareto-ranking methods seem sufficient and promising. Acknowledgments The research was supported by the research program No. MSM Transdisciplinary Research in the Area of Biomedical Engineering II of the CTU in Prague, sponsored by the Ministry of Education, Youth and Sports of the Czech Republic and CTU grant No References 1. Rabiner, R.: A tutorial on hidden markov models and selected applications in speech recognition. Proceedings of the IEEE 77(2) (1989) Kwong, S., Chau, C.: Analysis of paralell genetic algortihms on hmm based speech recognition system. In: ICASSP. (1997) Dempster, A., Laird, N., Rubin, D.: Maximum-likelihood from incomplete data via the em algorithm. Journal of Royal Statistics 39 (1977) Kennedy, J., Eberhart, R.: Particle swarm optimization. In: Proceedings International Conference on Neural Networks IEEE. Volume 4. (1995) X.Hu, R.C.Eberhart: Solving constrained nonlinear optimization problems with particle swarm optimization. In: Proceedings of the Sixth World Multiconference on Systems, Cybernetics and Informatics. (2002) 6. Paquet, U., p Engelbrecht, A.: A New Particle Swarm Optimiser for Linearly Constrained Optimization. In: Proceedings of the Congress on Evolutionary Computation 2003 (CEC 2003). Volume 1., Piscataway, New Jersey, Canberra, Australia, IEEE Service Center (2003)

Artificial Bee Colony (ABC) Optimization Algorithm for Solving Constrained Optimization Problems

Artificial Bee Colony (ABC) Optimization Algorithm for Solving Constrained Optimization Problems Artificial Bee Colony (ABC) Optimization Algorithm for Solving Constrained Optimization Problems Dervis Karaboga and Bahriye Basturk Erciyes University, Engineering Faculty, The Department of Computer

More information

An improved PID neural network controller for long time delay systems using particle swarm optimization algorithm

An improved PID neural network controller for long time delay systems using particle swarm optimization algorithm An improved PID neural network controller for long time delay systems using particle swarm optimization algorithm A. Lari, A. Khosravi and A. Alfi Faculty of Electrical and Computer Engineering, Noushirvani

More information

A MULTI-SWARM PARTICLE SWARM OPTIMIZATION WITH LOCAL SEARCH ON MULTI-ROBOT SEARCH SYSTEM

A MULTI-SWARM PARTICLE SWARM OPTIMIZATION WITH LOCAL SEARCH ON MULTI-ROBOT SEARCH SYSTEM A MULTI-SWARM PARTICLE SWARM OPTIMIZATION WITH LOCAL SEARCH ON MULTI-ROBOT SEARCH SYSTEM BAHAREH NAKISA, MOHAMMAD NAIM RASTGOO, MOHAMMAD FAIDZUL NASRUDIN, MOHD ZAKREE AHMAD NAZRI Department of Computer

More information

Modified Particle Swarm Optimization

Modified Particle Swarm Optimization Modified Particle Swarm Optimization Swati Agrawal 1, R.P. Shimpi 2 1 Aerospace Engineering Department, IIT Bombay, Mumbai, India, swati.agrawal@iitb.ac.in 2 Aerospace Engineering Department, IIT Bombay,

More information

Mobile Robot Path Planning in Static Environments using Particle Swarm Optimization

Mobile Robot Path Planning in Static Environments using Particle Swarm Optimization Mobile Robot Path Planning in Static Environments using Particle Swarm Optimization M. Shahab Alam, M. Usman Rafique, and M. Umer Khan Abstract Motion planning is a key element of robotics since it empowers

More information

Invariant Recognition of Hand-Drawn Pictograms Using HMMs with a Rotating Feature Extraction

Invariant Recognition of Hand-Drawn Pictograms Using HMMs with a Rotating Feature Extraction Invariant Recognition of Hand-Drawn Pictograms Using HMMs with a Rotating Feature Extraction Stefan Müller, Gerhard Rigoll, Andreas Kosmala and Denis Mazurenok Department of Computer Science, Faculty of

More information

A *69>H>N6 #DJGC6A DG C<>C::G>C<,8>:C8:H /DA 'D 2:6G, ()-"&"3 -"(' ( +-" " " % '.+ % ' -0(+$,

A *69>H>N6 #DJGC6A DG C<>C::G>C<,8>:C8:H /DA 'D 2:6G, ()-&3 -(' ( +-   % '.+ % ' -0(+$, The structure is a very important aspect in neural network design, it is not only impossible to determine an optimal structure for a given problem, it is even impossible to prove that a given structure

More information

Inductive Modelling of Temporal Sequences by Means of Self-organization

Inductive Modelling of Temporal Sequences by Means of Self-organization Inductive Modelling of Temporal Sequences by Means of Self-organization Jan Koutnik 1 1 Computational Intelligence Group, Dept. of Computer Science and Engineering, Faculty of Electrical Engineering, Czech

More information

CHAPTER 2 CONVENTIONAL AND NON-CONVENTIONAL TECHNIQUES TO SOLVE ORPD PROBLEM

CHAPTER 2 CONVENTIONAL AND NON-CONVENTIONAL TECHNIQUES TO SOLVE ORPD PROBLEM 20 CHAPTER 2 CONVENTIONAL AND NON-CONVENTIONAL TECHNIQUES TO SOLVE ORPD PROBLEM 2.1 CLASSIFICATION OF CONVENTIONAL TECHNIQUES Classical optimization methods can be classified into two distinct groups:

More information

Adaptively Choosing Neighbourhood Bests Using Species in a Particle Swarm Optimizer for Multimodal Function Optimization

Adaptively Choosing Neighbourhood Bests Using Species in a Particle Swarm Optimizer for Multimodal Function Optimization Adaptively Choosing Neighbourhood Bests Using Species in a Particle Swarm Optimizer for Multimodal Function Optimization Xiaodong Li School of Computer Science and Information Technology RMIT University,

More information

NOVEL HYBRID GENETIC ALGORITHM WITH HMM BASED IRIS RECOGNITION

NOVEL HYBRID GENETIC ALGORITHM WITH HMM BASED IRIS RECOGNITION NOVEL HYBRID GENETIC ALGORITHM WITH HMM BASED IRIS RECOGNITION * Prof. Dr. Ban Ahmed Mitras ** Ammar Saad Abdul-Jabbar * Dept. of Operation Research & Intelligent Techniques ** Dept. of Mathematics. College

More information

Artificial bee colony algorithm with multiple onlookers for constrained optimization problems

Artificial bee colony algorithm with multiple onlookers for constrained optimization problems Artificial bee colony algorithm with multiple onlookers for constrained optimization problems Milos Subotic Faculty of Computer Science University Megatrend Belgrade Bulevar umetnosti 29 SERBIA milos.subotic@gmail.com

More information

Small World Particle Swarm Optimizer for Global Optimization Problems

Small World Particle Swarm Optimizer for Global Optimization Problems Small World Particle Swarm Optimizer for Global Optimization Problems Megha Vora and T.T. Mirnalinee Department of Computer Science and Engineering S.S.N College of Engineering, Anna University, Chennai,

More information

International Journal of Digital Application & Contemporary research Website: (Volume 1, Issue 7, February 2013)

International Journal of Digital Application & Contemporary research Website:   (Volume 1, Issue 7, February 2013) Performance Analysis of GA and PSO over Economic Load Dispatch Problem Sakshi Rajpoot sakshirajpoot1988@gmail.com Dr. Sandeep Bhongade sandeepbhongade@rediffmail.com Abstract Economic Load dispatch problem

More information

Handling Multi Objectives of with Multi Objective Dynamic Particle Swarm Optimization

Handling Multi Objectives of with Multi Objective Dynamic Particle Swarm Optimization Handling Multi Objectives of with Multi Objective Dynamic Particle Swarm Optimization Richa Agnihotri #1, Dr. Shikha Agrawal #1, Dr. Rajeev Pandey #1 # Department of Computer Science Engineering, UIT,

More information

Hidden Markov Model for Sequential Data

Hidden Markov Model for Sequential Data Hidden Markov Model for Sequential Data Dr.-Ing. Michelle Karg mekarg@uwaterloo.ca Electrical and Computer Engineering Cheriton School of Computer Science Sequential Data Measurement of time series: Example:

More information

Particle Swarm Optimization

Particle Swarm Optimization Particle Swarm Optimization Gonçalo Pereira INESC-ID and Instituto Superior Técnico Porto Salvo, Portugal gpereira@gaips.inesc-id.pt April 15, 2011 1 What is it? Particle Swarm Optimization is an algorithm

More information

Adaptative Clustering Particle Swarm Optimization

Adaptative Clustering Particle Swarm Optimization Adaptative Clustering Particle Swarm Optimization Salomão S. Madeiro, Carmelo J. A. Bastos-Filho, Member, IEEE, and Fernando B. Lima Neto, Senior Member, IEEE, Elliackin M. N. Figueiredo Abstract The performance

More information

CHAPTER 6 HYBRID AI BASED IMAGE CLASSIFICATION TECHNIQUES

CHAPTER 6 HYBRID AI BASED IMAGE CLASSIFICATION TECHNIQUES CHAPTER 6 HYBRID AI BASED IMAGE CLASSIFICATION TECHNIQUES 6.1 INTRODUCTION The exploration of applications of ANN for image classification has yielded satisfactory results. But, the scope for improving

More information

Experimental Study on Bound Handling Techniques for Multi-Objective Particle Swarm Optimization

Experimental Study on Bound Handling Techniques for Multi-Objective Particle Swarm Optimization Experimental Study on Bound Handling Techniques for Multi-Objective Particle Swarm Optimization adfa, p. 1, 2011. Springer-Verlag Berlin Heidelberg 2011 Devang Agarwal and Deepak Sharma Department of Mechanical

More information

Constrained Single-Objective Optimization Using Particle Swarm Optimization

Constrained Single-Objective Optimization Using Particle Swarm Optimization 2006 IEEE Congress on Evolutionary Computation Sheraton Vancouver Wall Centre Hotel, Vancouver, BC, Canada July 16-21, 2006 Constrained Single-Objective Optimization Using Particle Swarm Optimization Karin

More information

Research Article Path Planning Using a Hybrid Evolutionary Algorithm Based on Tree Structure Encoding

Research Article Path Planning Using a Hybrid Evolutionary Algorithm Based on Tree Structure Encoding e Scientific World Journal, Article ID 746260, 8 pages http://dx.doi.org/10.1155/2014/746260 Research Article Path Planning Using a Hybrid Evolutionary Algorithm Based on Tree Structure Encoding Ming-Yi

More information

Hybrid Optimization Coupling Electromagnetism and Descent Search for Engineering Problems

Hybrid Optimization Coupling Electromagnetism and Descent Search for Engineering Problems Proceedings of the International Conference on Computational and Mathematical Methods in Science and Engineering, CMMSE 2008 13 17 June 2008. Hybrid Optimization Coupling Electromagnetism and Descent Search

More information

Machine Learning. Topic 5: Linear Discriminants. Bryan Pardo, EECS 349 Machine Learning, 2013

Machine Learning. Topic 5: Linear Discriminants. Bryan Pardo, EECS 349 Machine Learning, 2013 Machine Learning Topic 5: Linear Discriminants Bryan Pardo, EECS 349 Machine Learning, 2013 Thanks to Mark Cartwright for his extensive contributions to these slides Thanks to Alpaydin, Bishop, and Duda/Hart/Stork

More information

ARMA MODEL SELECTION USING PARTICLE SWARM OPTIMIZATION AND AIC CRITERIA. Mark S. Voss a b. and Xin Feng.

ARMA MODEL SELECTION USING PARTICLE SWARM OPTIMIZATION AND AIC CRITERIA. Mark S. Voss a b. and Xin Feng. Copyright 2002 IFAC 5th Triennial World Congress, Barcelona, Spain ARMA MODEL SELECTION USING PARTICLE SWARM OPTIMIZATION AND AIC CRITERIA Mark S. Voss a b and Xin Feng a Department of Civil and Environmental

More information

Inertia Weight. v i = ωv i +φ 1 R(0,1)(p i x i )+φ 2 R(0,1)(p g x i ) The new velocity update equation:

Inertia Weight. v i = ωv i +φ 1 R(0,1)(p i x i )+φ 2 R(0,1)(p g x i ) The new velocity update equation: Convergence of PSO The velocity update equation: v i = v i +φ 1 R(0,1)(p i x i )+φ 2 R(0,1)(p g x i ) for some values of φ 1 and φ 2 the velocity grows without bound can bound velocity to range [ V max,v

More information

Regularization of Evolving Polynomial Models

Regularization of Evolving Polynomial Models Regularization of Evolving Polynomial Models Pavel Kordík Dept. of Computer Science and Engineering, Karlovo nám. 13, 121 35 Praha 2, Czech Republic kordikp@fel.cvut.cz Abstract. Black box models such

More information

Meta- Heuristic based Optimization Algorithms: A Comparative Study of Genetic Algorithm and Particle Swarm Optimization

Meta- Heuristic based Optimization Algorithms: A Comparative Study of Genetic Algorithm and Particle Swarm Optimization 2017 2 nd International Electrical Engineering Conference (IEEC 2017) May. 19 th -20 th, 2017 at IEP Centre, Karachi, Pakistan Meta- Heuristic based Optimization Algorithms: A Comparative Study of Genetic

More information

PARALLEL PARTICLE SWARM OPTIMIZATION IN DATA CLUSTERING

PARALLEL PARTICLE SWARM OPTIMIZATION IN DATA CLUSTERING PARALLEL PARTICLE SWARM OPTIMIZATION IN DATA CLUSTERING YASIN ORTAKCI Karabuk University, Computer Engineering Department, Karabuk, Turkey E-mail: yasinortakci@karabuk.edu.tr Abstract Particle Swarm Optimization

More information

Simultaneous Perturbation Stochastic Approximation Algorithm Combined with Neural Network and Fuzzy Simulation

Simultaneous Perturbation Stochastic Approximation Algorithm Combined with Neural Network and Fuzzy Simulation .--- Simultaneous Perturbation Stochastic Approximation Algorithm Combined with Neural Networ and Fuzzy Simulation Abstract - - - - Keywords: Many optimization problems contain fuzzy information. Possibility

More information

Traffic Signal Control Based On Fuzzy Artificial Neural Networks With Particle Swarm Optimization

Traffic Signal Control Based On Fuzzy Artificial Neural Networks With Particle Swarm Optimization Traffic Signal Control Based On Fuzzy Artificial Neural Networks With Particle Swarm Optimization J.Venkatesh 1, B.Chiranjeevulu 2 1 PG Student, Dept. of ECE, Viswanadha Institute of Technology And Management,

More information

Assessing Particle Swarm Optimizers Using Network Science Metrics

Assessing Particle Swarm Optimizers Using Network Science Metrics Assessing Particle Swarm Optimizers Using Network Science Metrics Marcos A. C. Oliveira-Júnior, Carmelo J. A. Bastos-Filho and Ronaldo Menezes Abstract Particle Swarm Optimizers (PSOs) have been widely

More information

ModelStructureSelection&TrainingAlgorithmsfor an HMMGesture Recognition System

ModelStructureSelection&TrainingAlgorithmsfor an HMMGesture Recognition System ModelStructureSelection&TrainingAlgorithmsfor an HMMGesture Recognition System Nianjun Liu, Brian C. Lovell, Peter J. Kootsookos, and Richard I.A. Davis Intelligent Real-Time Imaging and Sensing (IRIS)

More information

GENETIC ALGORITHM VERSUS PARTICLE SWARM OPTIMIZATION IN N-QUEEN PROBLEM

GENETIC ALGORITHM VERSUS PARTICLE SWARM OPTIMIZATION IN N-QUEEN PROBLEM Journal of Al-Nahrain University Vol.10(2), December, 2007, pp.172-177 Science GENETIC ALGORITHM VERSUS PARTICLE SWARM OPTIMIZATION IN N-QUEEN PROBLEM * Azhar W. Hammad, ** Dr. Ban N. Thannoon Al-Nahrain

More information

Argha Roy* Dept. of CSE Netaji Subhash Engg. College West Bengal, India.

Argha Roy* Dept. of CSE Netaji Subhash Engg. College West Bengal, India. Volume 3, Issue 3, March 2013 ISSN: 2277 128X International Journal of Advanced Research in Computer Science and Software Engineering Research Paper Available online at: www.ijarcsse.com Training Artificial

More information

Neural Network Weight Selection Using Genetic Algorithms

Neural Network Weight Selection Using Genetic Algorithms Neural Network Weight Selection Using Genetic Algorithms David Montana presented by: Carl Fink, Hongyi Chen, Jack Cheng, Xinglong Li, Bruce Lin, Chongjie Zhang April 12, 2005 1 Neural Networks Neural networks

More information

Index Terms PSO, parallel computing, clustering, multiprocessor.

Index Terms PSO, parallel computing, clustering, multiprocessor. Parallel Particle Swarm Optimization in Data Clustering Yasin ORTAKCI Karabuk University, Computer Engineering Department, Karabuk, Turkey yasinortakci@karabuk.edu.tr Abstract Particle Swarm Optimization

More information

Using Gradient Descent Optimization for Acoustics Training from Heterogeneous Data

Using Gradient Descent Optimization for Acoustics Training from Heterogeneous Data Using Gradient Descent Optimization for Acoustics Training from Heterogeneous Data Martin Karafiát Λ, Igor Szöke, and Jan Černocký Brno University of Technology, Faculty of Information Technology Department

More information

FOREST PLANNING USING PSO WITH A PRIORITY REPRESENTATION

FOREST PLANNING USING PSO WITH A PRIORITY REPRESENTATION FOREST PLANNING USING PSO WITH A PRIORITY REPRESENTATION P.W. Brooks and W.D. Potter Institute for Artificial Intelligence, University of Georgia, USA Overview Background: (NIO Project 1 ) PSO -- GA --

More information

Clustering Sequences with Hidden. Markov Models. Padhraic Smyth CA Abstract

Clustering Sequences with Hidden. Markov Models. Padhraic Smyth CA Abstract Clustering Sequences with Hidden Markov Models Padhraic Smyth Information and Computer Science University of California, Irvine CA 92697-3425 smyth@ics.uci.edu Abstract This paper discusses a probabilistic

More information

Clustering Documents in Large Text Corpora

Clustering Documents in Large Text Corpora Clustering Documents in Large Text Corpora Bin He Faculty of Computer Science Dalhousie University Halifax, Canada B3H 1W5 bhe@cs.dal.ca http://www.cs.dal.ca/ bhe Yongzheng Zhang Faculty of Computer Science

More information

IMPROVING THE PARTICLE SWARM OPTIMIZATION ALGORITHM USING THE SIMPLEX METHOD AT LATE STAGE

IMPROVING THE PARTICLE SWARM OPTIMIZATION ALGORITHM USING THE SIMPLEX METHOD AT LATE STAGE IMPROVING THE PARTICLE SWARM OPTIMIZATION ALGORITHM USING THE SIMPLEX METHOD AT LATE STAGE Fang Wang, and Yuhui Qiu Intelligent Software and Software Engineering Laboratory, Southwest-China Normal University,

More information

Binary Differential Evolution Strategies

Binary Differential Evolution Strategies Binary Differential Evolution Strategies A.P. Engelbrecht, Member, IEEE G. Pampará Abstract Differential evolution has shown to be a very powerful, yet simple, population-based optimization approach. The

More information

Particle Swarm Optimization

Particle Swarm Optimization Dario Schor, M.Sc., EIT schor@ieee.org Space Systems Department Magellan Aerospace Winnipeg Winnipeg, Manitoba 1 of 34 Optimization Techniques Motivation Optimization: Where, min x F(x), subject to g(x)

More information

Probabilistic Facial Feature Extraction Using Joint Distribution of Location and Texture Information

Probabilistic Facial Feature Extraction Using Joint Distribution of Location and Texture Information Probabilistic Facial Feature Extraction Using Joint Distribution of Location and Texture Information Mustafa Berkay Yilmaz, Hakan Erdogan, Mustafa Unel Sabanci University, Faculty of Engineering and Natural

More information

A Novel Hybrid Self Organizing Migrating Algorithm with Mutation for Global Optimization

A Novel Hybrid Self Organizing Migrating Algorithm with Mutation for Global Optimization International Journal of Soft Computing and Engineering (IJSCE) ISSN: 2231-2307, Volume-3, Issue-6, January 2014 A Novel Hybrid Self Organizing Migrating Algorithm with Mutation for Global Optimization

More information

Estimation of Hidden Markov Models Parameters using Differential Evolution

Estimation of Hidden Markov Models Parameters using Differential Evolution Estimation of Hidden Markov Models Parameters using Differential Evolution Ângela A. R. Sá 1, Adriano O. Andrade 1, Alcimar B. Soares 1 and Slawomir J. Nasuto 2 Abstract. Hidden Markov Models (HMMs) have

More information

Feature weighting using particle swarm optimization for learning vector quantization classifier

Feature weighting using particle swarm optimization for learning vector quantization classifier Journal of Physics: Conference Series PAPER OPEN ACCESS Feature weighting using particle swarm optimization for learning vector quantization classifier To cite this article: A Dongoran et al 2018 J. Phys.:

More information

THREE PHASE FAULT DIAGNOSIS BASED ON RBF NEURAL NETWORK OPTIMIZED BY PSO ALGORITHM

THREE PHASE FAULT DIAGNOSIS BASED ON RBF NEURAL NETWORK OPTIMIZED BY PSO ALGORITHM THREE PHASE FAULT DIAGNOSIS BASED ON RBF NEURAL NETWORK OPTIMIZED BY PSO ALGORITHM M. Sivakumar 1 and R. M. S. Parvathi 2 1 Anna University, Tamilnadu, India 2 Sengunthar College of Engineering, Tamilnadu,

More information

A Visualization Tool to Improve the Performance of a Classifier Based on Hidden Markov Models

A Visualization Tool to Improve the Performance of a Classifier Based on Hidden Markov Models A Visualization Tool to Improve the Performance of a Classifier Based on Hidden Markov Models Gleidson Pegoretti da Silva, Masaki Nakagawa Department of Computer and Information Sciences Tokyo University

More information

Study on the Development of Complex Network for Evolutionary and Swarm based Algorithms

Study on the Development of Complex Network for Evolutionary and Swarm based Algorithms Study on the Development of Complex Network for Evolutionary and Swarm based Algorithms 1 Roman Senkerik, 2 Ivan Zelinka, 1 Michal Pluhacek and 1 Adam Viktorin 1 Tomas Bata University in Zlin, Faculty

More information

Reference Point-Based Particle Swarm Optimization Using a Steady-State Approach

Reference Point-Based Particle Swarm Optimization Using a Steady-State Approach Reference Point-Based Particle Swarm Optimization Using a Steady-State Approach Richard Allmendinger,XiaodongLi 2,andJürgen Branke University of Karlsruhe, Institute AIFB, Karlsruhe, Germany 2 RMIT University,

More information

Discrete Multi-Valued Particle Swarm Optimization

Discrete Multi-Valued Particle Swarm Optimization Discrete Multi-d Particle Swarm Optimization Jim Pugh and Alcherio Martinoli Swarm-Intelligent Systems Group École Polytechnique Fédérale de Lausanne 05 Lausanne, Switzerland Email: {jim.pugh,alcherio.martinoli}@epfl.ch

More information

Parallel Neural Network Training with OpenCL

Parallel Neural Network Training with OpenCL Parallel Neural Network Training with OpenCL Nenad Krpan, Domagoj Jakobović Faculty of Electrical Engineering and Computing Unska 3, Zagreb, Croatia Email: nenadkrpan@gmail.com, domagoj.jakobovic@fer.hr

More information

CLASSIFICATION WITH RADIAL BASIS AND PROBABILISTIC NEURAL NETWORKS

CLASSIFICATION WITH RADIAL BASIS AND PROBABILISTIC NEURAL NETWORKS CLASSIFICATION WITH RADIAL BASIS AND PROBABILISTIC NEURAL NETWORKS CHAPTER 4 CLASSIFICATION WITH RADIAL BASIS AND PROBABILISTIC NEURAL NETWORKS 4.1 Introduction Optical character recognition is one of

More information

Chapter 14 Global Search Algorithms

Chapter 14 Global Search Algorithms Chapter 14 Global Search Algorithms An Introduction to Optimization Spring, 2015 Wei-Ta Chu 1 Introduction We discuss various search methods that attempts to search throughout the entire feasible set.

More information

Particle Swarm Optimization Based Learning Method for Process Neural Networks

Particle Swarm Optimization Based Learning Method for Process Neural Networks Particle Swarm Optimization Based Learning Method for Process Neural Networks Kun Liu, Ying Tan, and Xingui He Key Laboratory of Machine Perception, Ministry of Education, Peking University, Beijing 100871,

More information

CHAPTER 5 OPTIMAL CLUSTER-BASED RETRIEVAL

CHAPTER 5 OPTIMAL CLUSTER-BASED RETRIEVAL 85 CHAPTER 5 OPTIMAL CLUSTER-BASED RETRIEVAL 5.1 INTRODUCTION Document clustering can be applied to improve the retrieval process. Fast and high quality document clustering algorithms play an important

More information

Particle Swarm Optimization applied to Pattern Recognition

Particle Swarm Optimization applied to Pattern Recognition Particle Swarm Optimization applied to Pattern Recognition by Abel Mengistu Advisor: Dr. Raheel Ahmad CS Senior Research 2011 Manchester College May, 2011-1 - Table of Contents Introduction... - 3 - Objectives...

More information

A New Improved Hybridized K-MEANS Clustering Algorithm with Improved PCA Optimized with PSO for High Dimensional Data Set

A New Improved Hybridized K-MEANS Clustering Algorithm with Improved PCA Optimized with PSO for High Dimensional Data Set International Journal of Soft Computing and Engineering (IJSCE) ISSN: 2231-2307, Volume-2, Issue-2, May 2012 A New Improved Hybridized K-MEANS Clustering Algorithm with Improved PCA Optimized with PSO

More information

A NEW APPROACH TO SOLVE ECONOMIC LOAD DISPATCH USING PARTICLE SWARM OPTIMIZATION

A NEW APPROACH TO SOLVE ECONOMIC LOAD DISPATCH USING PARTICLE SWARM OPTIMIZATION A NEW APPROACH TO SOLVE ECONOMIC LOAD DISPATCH USING PARTICLE SWARM OPTIMIZATION Manjeet Singh 1, Divesh Thareja 2 1 Department of Electrical and Electronics Engineering, Assistant Professor, HCTM Technical

More information

Stephen Scott.

Stephen Scott. 1 / 33 sscott@cse.unl.edu 2 / 33 Start with a set of sequences In each column, residues are homolgous Residues occupy similar positions in 3D structure Residues diverge from a common ancestral residue

More information

Person Authentication from Video of Faces: A Behavioral and Physiological Approach Using Pseudo Hierarchical Hidden Markov Models

Person Authentication from Video of Faces: A Behavioral and Physiological Approach Using Pseudo Hierarchical Hidden Markov Models Person Authentication from Video of Faces: A Behavioral and Physiological Approach Using Pseudo Hierarchical Hidden Markov Models Manuele Bicego 1, Enrico Grosso 1, and Massimo Tistarelli 2 1 DEIR - University

More information

Automatic differentiation based for particle swarm optimization steepest descent direction

Automatic differentiation based for particle swarm optimization steepest descent direction International Journal of Advances in Intelligent Informatics ISSN: 2442-6571 Vol 1, No 2, July 2015, pp. 90-97 90 Automatic differentiation based for particle swarm optimization steepest descent direction

More information

DERIVATIVE-FREE OPTIMIZATION

DERIVATIVE-FREE OPTIMIZATION DERIVATIVE-FREE OPTIMIZATION Main bibliography J.-S. Jang, C.-T. Sun and E. Mizutani. Neuro-Fuzzy and Soft Computing: A Computational Approach to Learning and Machine Intelligence. Prentice Hall, New Jersey,

More information

Theoretical Concepts of Machine Learning

Theoretical Concepts of Machine Learning Theoretical Concepts of Machine Learning Part 2 Institute of Bioinformatics Johannes Kepler University, Linz, Austria Outline 1 Introduction 2 Generalization Error 3 Maximum Likelihood 4 Noise Models 5

More information

Automatic Speech Recognition Using Support Vector Machine and Particle Swarm Optimization

Automatic Speech Recognition Using Support Vector Machine and Particle Swarm Optimization Automatic Speech Recognition Using Support Vector Machine and Particle Swarm Optimization Gracieth Cavalcanti Batista Federal Institute of Maranhao Student of Electrical Engineering Sao Luis, Maranhao,

More information

Improving Tree-Based Classification Rules Using a Particle Swarm Optimization

Improving Tree-Based Classification Rules Using a Particle Swarm Optimization Improving Tree-Based Classification Rules Using a Particle Swarm Optimization Chi-Hyuck Jun *, Yun-Ju Cho, and Hyeseon Lee Department of Industrial and Management Engineering Pohang University of Science

More information

Constrained Classification of Large Imbalanced Data

Constrained Classification of Large Imbalanced Data Constrained Classification of Large Imbalanced Data Martin Hlosta, R. Stríž, J. Zendulka, T. Hruška Brno University of Technology, Faculty of Information Technology Božetěchova 2, 612 66 Brno ihlosta@fit.vutbr.cz

More information

Note Set 4: Finite Mixture Models and the EM Algorithm

Note Set 4: Finite Mixture Models and the EM Algorithm Note Set 4: Finite Mixture Models and the EM Algorithm Padhraic Smyth, Department of Computer Science University of California, Irvine Finite Mixture Models A finite mixture model with K components, for

More information

SwarmOps for Java. Numeric & Heuristic Optimization Source-Code Library for Java The Manual Revision 1.0

SwarmOps for Java. Numeric & Heuristic Optimization Source-Code Library for Java The Manual Revision 1.0 Numeric & Heuristic Optimization Source-Code Library for Java The Manual Revision 1.0 By Magnus Erik Hvass Pedersen June 2011 Copyright 2009-2011, all rights reserved by the author. Please see page 4 for

More information

CHAPTER 6 ORTHOGONAL PARTICLE SWARM OPTIMIZATION

CHAPTER 6 ORTHOGONAL PARTICLE SWARM OPTIMIZATION 131 CHAPTER 6 ORTHOGONAL PARTICLE SWARM OPTIMIZATION 6.1 INTRODUCTION The Orthogonal arrays are helpful in guiding the heuristic algorithms to obtain a good solution when applied to NP-hard problems. This

More information

CS839: Probabilistic Graphical Models. Lecture 10: Learning with Partially Observed Data. Theo Rekatsinas

CS839: Probabilistic Graphical Models. Lecture 10: Learning with Partially Observed Data. Theo Rekatsinas CS839: Probabilistic Graphical Models Lecture 10: Learning with Partially Observed Data Theo Rekatsinas 1 Partially Observed GMs Speech recognition 2 Partially Observed GMs Evolution 3 Partially Observed

More information

Model Parameter Estimation

Model Parameter Estimation Model Parameter Estimation Shan He School for Computational Science University of Birmingham Module 06-23836: Computational Modelling with MATLAB Outline Outline of Topics Concepts about model parameter

More information

Application of Improved Discrete Particle Swarm Optimization in Logistics Distribution Routing Problem

Application of Improved Discrete Particle Swarm Optimization in Logistics Distribution Routing Problem Available online at www.sciencedirect.com Procedia Engineering 15 (2011) 3673 3677 Advanced in Control Engineeringand Information Science Application of Improved Discrete Particle Swarm Optimization in

More information

Use of Extreme Value Statistics in Modeling Biometric Systems

Use of Extreme Value Statistics in Modeling Biometric Systems Use of Extreme Value Statistics in Modeling Biometric Systems Similarity Scores Two types of matching: Genuine sample Imposter sample Matching scores Enrolled sample 0.95 0.32 Probability Density Decision

More information

Particle Swarm Optimization to Solve Optimization Problems

Particle Swarm Optimization to Solve Optimization Problems Particle Swarm Optimization to Solve Optimization Problems Gregorio Toscano-Pulido and Carlos A. Coello Coello Evolutionary Computation Group at CINVESTAV-IPN (EVOCINV) Electrical Eng. Department, Computer

More information

Effect of the PSO Topologies on the Performance of the PSO-ELM

Effect of the PSO Topologies on the Performance of the PSO-ELM 2012 Brazilian Symposium on Neural Networks Effect of the PSO Topologies on the Performance of the PSO-ELM Elliackin M. N. Figueiredo and Teresa B. Ludermir Center of Informatics Federal University of

More information

HMM-Based Handwritten Amharic Word Recognition with Feature Concatenation

HMM-Based Handwritten Amharic Word Recognition with Feature Concatenation 009 10th International Conference on Document Analysis and Recognition HMM-Based Handwritten Amharic Word Recognition with Feature Concatenation Yaregal Assabie and Josef Bigun School of Information Science,

More information

Random projection for non-gaussian mixture models

Random projection for non-gaussian mixture models Random projection for non-gaussian mixture models Győző Gidófalvi Department of Computer Science and Engineering University of California, San Diego La Jolla, CA 92037 gyozo@cs.ucsd.edu Abstract Recently,

More information

A New Discrete Binary Particle Swarm Optimization based on Learning Automata

A New Discrete Binary Particle Swarm Optimization based on Learning Automata A New Discrete Binary Particle Swarm Optimization based on Learning Automata R. Rastegar M. R. Meybodi K. Badie Soft Computing Lab Soft Computing Lab Information Technology Computer Eng. Department Computer

More information

Cell-to-switch assignment in. cellular networks. barebones particle swarm optimization

Cell-to-switch assignment in. cellular networks. barebones particle swarm optimization Cell-to-switch assignment in cellular networks using barebones particle swarm optimization Sotirios K. Goudos a), Konstantinos B. Baltzis, Christos Bachtsevanidis, and John N. Sahalos RadioCommunications

More information

DISTANCE EVALUATED SIMULATED KALMAN FILTER FOR COMBINATORIAL OPTIMIZATION PROBLEMS

DISTANCE EVALUATED SIMULATED KALMAN FILTER FOR COMBINATORIAL OPTIMIZATION PROBLEMS DISTANCE EVALUATED SIMULATED KALMAN FILTER FOR COMBINATORIAL OPTIMIZATION PROBLEMS Zulkifli Md Yusof 1, Zuwairie Ibrahim 1, Ismail Ibrahim 1, Kamil Zakwan Mohd Azmi 1, Nor Azlina Ab Aziz 2, Nor Hidayati

More information

PARTICLE SWARM OPTIMIZATION (PSO)

PARTICLE SWARM OPTIMIZATION (PSO) PARTICLE SWARM OPTIMIZATION (PSO) J. Kennedy and R. Eberhart, Particle Swarm Optimization. Proceedings of the Fourth IEEE Int. Conference on Neural Networks, 1995. A population based optimization technique

More information

Hybrid Particle Swarm-Based-Simulated Annealing Optimization Techniques

Hybrid Particle Swarm-Based-Simulated Annealing Optimization Techniques Hybrid Particle Swarm-Based-Simulated Annealing Optimization Techniques Nasser Sadati Abstract Particle Swarm Optimization (PSO) algorithms recently invented as intelligent optimizers with several highly

More information

Simplifying Handwritten Characters Recognition Using a Particle Swarm Optimization Approach

Simplifying Handwritten Characters Recognition Using a Particle Swarm Optimization Approach ISSN 2286-4822, www.euacademic.org IMPACT FACTOR: 0.485 (GIF) Simplifying Handwritten Characters Recognition Using a Particle Swarm Optimization Approach MAJIDA ALI ABED College of Computers Sciences and

More information

Feature scaling in support vector data description

Feature scaling in support vector data description Feature scaling in support vector data description P. Juszczak, D.M.J. Tax, R.P.W. Duin Pattern Recognition Group, Department of Applied Physics, Faculty of Applied Sciences, Delft University of Technology,

More information

Particle Swarm Optimization Artificial Bee Colony Chain (PSOABCC): A Hybrid Meteahuristic Algorithm

Particle Swarm Optimization Artificial Bee Colony Chain (PSOABCC): A Hybrid Meteahuristic Algorithm Particle Swarm Optimization Artificial Bee Colony Chain (PSOABCC): A Hybrid Meteahuristic Algorithm Oğuz Altun Department of Computer Engineering Yildiz Technical University Istanbul, Turkey oaltun@yildiz.edu.tr

More information

Flexible-Hybrid Sequential Floating Search in Statistical Feature Selection

Flexible-Hybrid Sequential Floating Search in Statistical Feature Selection Flexible-Hybrid Sequential Floating Search in Statistical Feature Selection Petr Somol 1,2, Jana Novovičová 1,2, and Pavel Pudil 2,1 1 Dept. of Pattern Recognition, Institute of Information Theory and

More information

An Island Based Hybrid Evolutionary Algorithm for Optimization

An Island Based Hybrid Evolutionary Algorithm for Optimization An Island Based Hybrid Evolutionary Algorithm for Optimization Changhe Li and Shengxiang Yang Department of Computer Science, University of Leicester University Road, Leicester LE1 7RH, UK {cl160,s.yang}@mcs.le.ac.uk

More information

QUANTUM BASED PSO TECHNIQUE FOR IMAGE SEGMENTATION

QUANTUM BASED PSO TECHNIQUE FOR IMAGE SEGMENTATION International Journal of Computer Engineering and Applications, Volume VIII, Issue I, Part I, October 14 QUANTUM BASED PSO TECHNIQUE FOR IMAGE SEGMENTATION Shradha Chawla 1, Vivek Panwar 2 1 Department

More information

Variable Neighborhood Particle Swarm Optimization for Multi-objective Flexible Job-Shop Scheduling Problems

Variable Neighborhood Particle Swarm Optimization for Multi-objective Flexible Job-Shop Scheduling Problems Variable Neighborhood Particle Swarm Optimization for Multi-objective Flexible Job-Shop Scheduling Problems Hongbo Liu 1,2,AjithAbraham 3,1, Okkyung Choi 3,4, and Seong Hwan Moon 4 1 School of Computer

More information

ECE521: Week 11, Lecture March 2017: HMM learning/inference. With thanks to Russ Salakhutdinov

ECE521: Week 11, Lecture March 2017: HMM learning/inference. With thanks to Russ Salakhutdinov ECE521: Week 11, Lecture 20 27 March 2017: HMM learning/inference With thanks to Russ Salakhutdinov Examples of other perspectives Murphy 17.4 End of Russell & Norvig 15.2 (Artificial Intelligence: A Modern

More information

Image Segmentation Using Iterated Graph Cuts BasedonMulti-scaleSmoothing

Image Segmentation Using Iterated Graph Cuts BasedonMulti-scaleSmoothing Image Segmentation Using Iterated Graph Cuts BasedonMulti-scaleSmoothing Tomoyuki Nagahashi 1, Hironobu Fujiyoshi 1, and Takeo Kanade 2 1 Dept. of Computer Science, Chubu University. Matsumoto 1200, Kasugai,

More information

Fuzzy Ant Clustering by Centroid Positioning

Fuzzy Ant Clustering by Centroid Positioning Fuzzy Ant Clustering by Centroid Positioning Parag M. Kanade and Lawrence O. Hall Computer Science & Engineering Dept University of South Florida, Tampa FL 33620 @csee.usf.edu Abstract We

More information

DEMO: Differential Evolution for Multiobjective Optimization

DEMO: Differential Evolution for Multiobjective Optimization DEMO: Differential Evolution for Multiobjective Optimization Tea Robič and Bogdan Filipič Department of Intelligent Systems, Jožef Stefan Institute, Jamova 39, SI-1000 Ljubljana, Slovenia tea.robic@ijs.si

More information

ARTIFICIAL INTELLIGENCE (CSCU9YE ) LECTURE 5: EVOLUTIONARY ALGORITHMS

ARTIFICIAL INTELLIGENCE (CSCU9YE ) LECTURE 5: EVOLUTIONARY ALGORITHMS ARTIFICIAL INTELLIGENCE (CSCU9YE ) LECTURE 5: EVOLUTIONARY ALGORITHMS Gabriela Ochoa http://www.cs.stir.ac.uk/~goc/ OUTLINE Optimisation problems Optimisation & search Two Examples The knapsack problem

More information

Hybrid PSO-SA algorithm for training a Neural Network for Classification

Hybrid PSO-SA algorithm for training a Neural Network for Classification Hybrid PSO-SA algorithm for training a Neural Network for Classification Sriram G. Sanjeevi 1, A. Naga Nikhila 2,Thaseem Khan 3 and G. Sumathi 4 1 Associate Professor, Dept. of CSE, National Institute

More information

Constraint Handling. Fernando Lobo. University of Algarve

Constraint Handling. Fernando Lobo. University of Algarve Constraint Handling Fernando Lobo University of Algarve Outline Introduction Penalty methods Approach based on tournament selection Decoders Repair algorithms Constraint-preserving operators Introduction

More information

Tracking Changing Extrema with Particle Swarm Optimizer

Tracking Changing Extrema with Particle Swarm Optimizer Tracking Changing Extrema with Particle Swarm Optimizer Anthony Carlisle Department of Mathematical and Computer Sciences, Huntingdon College antho@huntingdon.edu Abstract The modification of the Particle

More information