Digbalay Bose

Optimal filter design using an improved artificial bee colony algorithm

The domain of analog filter design revolves around the selection of proper values of the circuit components from a possible set of values manufactured keeping in mind the associated cost overhead. Normal design procedures result in a set of values for the discrete components that do not match with the preferred set of values. This results in the selection of approximated values that cause error in the associated design process. An optimal solution to the design problem would include selection of the best possible set of components from the numerous possible combinations. The search procedure for such an optimal solution necessitates the usage of Evolutionary Computation (EC) as a potential tool for determining the best possible set of circuit components. Recently algorithms based on Swarm Intelligence (SI) have gained prominence due to the underlying focus on collective intelligent behavior. In this paper a novel hybrid variant of a swarm-based metaheuristics called Artificial Bee Colony (ABC) algorithm is proposed and shall be referred to as CRbABC_Dt (Collective Resource-based ABC with Decentralized tasking) and it incorporates the idea of decentralization of attraction from super-fit members along with neighborhood information and wider exploration of search space. Two separate filter design instances have been tested using CRbABC_Dt algorithm and the results obtained are compared with several competitive state-of-the-art optimizing algorithms. All the components considered in the design are selected from standard series and the resulting deviation from the idealized design procedure has been investigated. Additional empirical experimentation has also been included based on the benchmarking problems proposed for the CEC 2013 Special Session & Competition on Real-Parameter Single Objective Optimization.

Migrating forager population in a multi-population Artificial Bee Colony algorithm with modified perturbation schemes

Swarm Intelligent algorithms focus on imbibing the collective intelligence of a group of simple agents that can work together as a unit. This research article focus on a recently proposed swarm-based metaheuristic called the Artificial Bee Colony (ABC) algorithm and suggests modifications to the algorithmic framework in order to enhance its performance. The proposed ABC variant shall be referred to as MsABC_Fm (Multi swarm Artificial Bee Colony with Forager migration). MsABC_Fm maintains multiple swarm populations that apply different perturbation strategies and gradually migration of the population from worse performing strategy to the better mode of perturbation is promoted. To evaluate the performance of the algorithm, we conduct comparative study involving 8 algorithms and test the problems on 25 benchmark problems proposed in the Special Session on IEEE Congress on Evolutionary Competition 2005. The superiority of the MsABC_Fm approach is also highlighted statistically.

A strategy pool adaptive artificial bee colony algorithm for dynamic environment through multi-population approach

Swarm Intelligence is based on developing metaheuristics that are modeled on certain life-sustaining principles exhibited by the biotic components of the ecosystem. There has been a surge in interest for nature inspired computing for devising more efficient models that can find solution to real-world problems using minimal resources at disposal. In this paper, an enhanced version of Artificial Bee Colony algorithm have been proposed that takes on the task of finding the optimal solution in a continuously changing (dynamic) solution space by incorporating a pool of varied perturbation strategies that operate on a multi-population group and synergizing the strategy pool with a set of diversity-inclusion techniques that help to maintain population diversity.

Decomposition-Based Evolutionary Multi-Objective Optimization Approach to the Design of Concentric Circular Antenna Arrays

We investigate the design of Concentric Circular Antenna Arrays (CCAAs) with ‚=2 uniform inter-element spacing, non-uniform radial separation, and non-uniform excitation across difierent rings, from the perspective of Multi-objective Optimization (MO). Unlike the existing single-objective design approaches that try to minimize a weighted sum of the design objectives like Side Lobe Level (SLL) and principal lobe Beam-Width (BW), we treat these two objectives individually and use Multiobjective Evolutionary Algorithm based on Decomposition (MOEA/D) with Difierential Evolution (DE), called MOEA/D-DE, to achieve the best tradeofi between the two objectives. Unlike the single-objective approaches, the MO approach provides greater ∞exibility in the design by yielding a set of equivalent flnal (non- dominated) solutions, from which the user can choose one that attains a suitable trade-ofi margin as per requirements. We illustrate that the best compromise solution attained by MOEA/D-DE can comfortably outperform state-of-the-art variants of single-objective algorithms like Particle Swarm Optimization (PSO) and Difierential Evolution. In addition, we compared the results obtained by MOEA/D-DE with those obtained by one of the most widely used MO algorithm called NSGA-2 and a multi-objective DE variant, on the basis of the R- indicator, hypervolume indicator, and quality of the best trade- ofi solutions obtained. Our simulation results clearly indicate the superiority of the design based on MOEA/D-DE.

Circular antenna array design using novel perturbation based artificial bee colony algorithm

The field of optimization is abundant with algorithms, which are inspired from nature based phenomena. The increasing popularity of such algorithms stems from their applications in real life situations. Here in this article a real life problem in the form of the design of circular antenna array has been discussed. The design of the antenna array is based on the application of a novel variant of Artificial Bee Colony Algorithm using selective neighborhood called sNABC. We use a neighborhood based perturbation on the basis of Euclidean distance and fitness of individuals are used for obtaining minimum side lobe levels, maximum directivity and appropriate null control. To illustrate the effectiveness of our design procedure, the results have been compared with several existing algorithms like DE, ABC and PSO.

A clustering particle based artificial bee colony algorithm for dynamic environment

Modern day real world applications present us challenging instances where the system needs to adapt to a changing environment without any sacrifice in its optimality. This led researchers to lay the foundations of dynamic problems in the field of optimization. Literature shows different approaches undertaken to tackle the problem of dynamic environment including techniques like diversity scheme, memory, multi-population scheme etc. In this paper we have proposed a hybrid scheme by combining k-means clustering technique with modified Artificial Bee Colony (ABC) algorithm as the base optimizer and it is expected that the clusters locate the optima in the problem. Experimental benchmark set that appeared in IEEE CEC 2009 has been used as test-bed and our ClPABC (Clustering Particle ABC) algorithm is compared against 4 state-of-the-art algorithms. The results show the superiority of our ClPABC approach on dynamic environment.

Cooperative co-evolutionary teaching-learning based algorithm with a modified exploration strategy for large scale global optimization

Evolutionary Algorithms, inspired from the Darwinian theory on evolution of species, are heuristic method for solving difficult unimodal and multimodal functions. But the ultimate disadvantage of those Evolutionary Algorithms is premature convergence, i.e. trapping in a local optimum due to poor exploration strategy. In case of High Dimensional problems, there are huge chances of convergence prematurely due to the large search space, which grows exponentially with the increase of dimension of the problem. In this paper a modified Teaching-Learning-Based technique is used to investigate the effectiveness of different cooperative co-evolutionary framework for solving high dimensional problems.

A selective teaching-learning based niching technique with local diversification strategy

Real world problems present instances where more than one optimal solution can be obtained for a system under consideration so as to switch between them without considerably affecting efficiency. In such instances the idea of niching provides a solution. In this paper we propose a swarm-based niching technique that enhances diversity by Teaching and Learning strategy that adapts to the local neighbourhood by controlled exploitation and the knowledge learned helps to preserve population diversity. Our algorithm, imitates the local-explorative swarm behaviour to hover around local sites in groups, exploiting the peaks with high degree of accuracy, is called TLB-lDS (Teaching-Learning Based Optimization with Local Diversification Strategy), without using any niching parameter. TLB-lDS algorithm is compared against sophisticated niching algorithms tested on a set of standard numerical benchmarks.

Synchronizing Differential Evolution with a modified affinity-based mutation framework

Differential Evolution is a stochastic, population-based optimization algorithm that has gained wide popularity these days for solving multi-modal, non-smooth, non-convex, and ill-behaved optimization problems. In this research article, we propose a restrictive mutation strategy that helps to probabilistically select individuals for mutation based on the information conveyed by neighboring individuals. The strategy is to develop a generalized approach that can restrict the stochastic selection by a more guided technique depending on distribution of adjacent individuals. Our approach takes into account both the proximity and the gradient estimation of the neighboring members of an individual to compute the selection probability. This framework can be easily integrated with basic DE and its state-of-the-art variants with minor changes. Experimental analysis reveals the superiority of our framework over the original variants when tested on the real parameter benchmark problems proposed in the IEEE Congress on Evolutionary Computation 2005 competition.

Clustering using vector membership: An extension of the Fuzzy C-Means algorithm

Clustering is an important facet of explorative data mining and finds extensive use in several fields. In this paper, we propose an extension of the classical Fuzzy C-Means clustering algorithm. The proposed algorithm, abbreviated as VFC, adopts a multi-dimensional membership vector for each data point instead of the traditional, scalar membership value defined in the original algorithm. The membership vector for each point is obtained by considering each feature of that point separately and obtaining individual membership values for the same. We also propose an algorithm to efficiently allocate the initial cluster centers close to the actual centers, so as to facilitate rapid convergence. Further, we propose a scheme to achieve crisp clustering using the VFC algorithm. The proposed, novel clustering scheme has been tested on two standard data sets in order to analyze its performance. We also examine the efficacy of the proposed scheme by analyzing its performance on image segmentation examples and comparing it with the classical Fuzzy C-means clustering algorithm.