Searching for memory in artificial immune system

Size: px
Start display at page:

Download "Searching for memory in artificial immune system"

Transcription

1 Searching for memory in artificial immune system Krzysztof Trojanowski 1), Sławomir T. Wierzchoń 1,2 1) Institute of Computer Science, Polish Academy of Sciences Warszwa, ul. Ordona ) Department of Computer Science, Białystok Technical University Białystok, ul. Wiejska 45 a Abstract: In this paper an idea of the artificial immune system was used to design an algorithm for non-stationary function optimization. It was demonstrated that in the case of periodic function changes the algorithm constructively builds and uses immune memory. This result was contrasted with cases when no periodic changes occur. Further, an attempt towards the identification of optimal partitioning of the antibodies population into antibodies subjected clonal selection and programmed death of cells (apoptosis) has been done. Keywords: Artificial Immune Systems, Clonal Selection, Apoptosis, Immune Memory, Nonstationary Optimization 1. Introduction Genetic algorithm (GA) a probabilistic algorithm solving a wide range of problems is in a sense a valuable instantiation of the General Problem Solver (GPS), [1], the dream of pioneers of Artificial Intelligence. It was observed by Nowell and Simon in the late fifties that the binary strings can be used for representing numbers as well as more complicated symbols. This observation gave an impulse to construct a system being able to solve general class of problems in a way similar to human problem solving. While the idea of GPS has failed (although expert systems can be viewed as its specialization), GAs still prove their usefulness and successful applications stimulate their development. From an abstract point of view classical GA can be treated as a string evolver. Using genetic operators of crossover and mutation it modifies the population of chromosomes, represented by binary strings, to produce at least one string that is as close as possible to a target (although unknown to the algorithm) string exploiting so-called fitness function as the only information about the

2 degree of closeness. To explain successfulness of such a search strategy, Holland formulated Schema Theorem, [2], according to which the GA assigns exponentially increasing number of trials to the observed best parts of the search space, what results in a convergence to the target string. However, this convergence is not always advantageous. As stated by Gaspar and Collard in [3], in fact it contradicts basic principle of natural evolution, where a great diversity of different species is observed. In other words, GA cannot maintain sufficient population diversity what results in its poor behavior when solving multimodal or time-dependent optimization problems. Efficiency of GA hardly depends on the trade-off between its explorative and exploitative abilities. When exploitation dominates exploration, the algorithm finds suboptimal solutions. Otherwise the algorithm vast computer resources exploring uninterested regions of the search space. To gain the appropriate tradeoff, a number of selection strategies has been proposed. Recently, a new biologically inspired technique, so-called artificial immune systems (AIS), have been proposed to overcome the problem with finding appropriate trade-off. The learning/adaptive mechanisms used by the vertebrate immune system allows continuous generation of new species of so-called antibodies responsible for detection and destruction of foreign molecules, called antigens or pathogens. Particularly these mechanisms, described in Section 2, appear to be useful in solving multimodal, [4], and time-dependent, [3], [5], optimization problems. In this paper we trace the emergence of the immune memory and its role in solving time-dependent optimization tasks. The paper is organized as follows. Section 2 introduces basic mechanisms used by the vertebrate immune system. The immune algorithm based on these mechanisms is described in Section 3. The environment designed for our experiments is presented in Section 4 and results of these experiments are described in Section 5. Section 6 concludes the paper. 2. Immune system While GA refers to the rules of Darwinian evolution relying upon introduction of permanent improvements in phenotypic properties of subsequent generations of living organisms, AIS refer to the mechanisms used by the adaptive layer of the immune system. The main actors of this system are lymphocytes or white cells of blood. We distinguish two important types of lymphocytes: B- lymphocytes (or B-cells for short) produced in bone marrow, and T-cells produced in thymus. Both the types differ in the roles fulfilled in the defense process. Roughly speaking T-cells are responsible for the detection between self and nonself substances while B-cells are involved in the production of so-called antibodies. Using military metaphor we can treat B-cell as a group of commandos equipped with selected weapon while T-cells are their commanders. From a computer science standpoint the mechanisms governing T-cells are used in designing novelty-detection systems (e.g. computer viruses detection) and the mechanisms governing B-cells are used in designing data analysis systems or optimization algorithms. Thus in the sequel we will focus on B-cells only. 2

3 B-lymphocyte is a monoclonal cell with about 1 5 receptors (antibodies) located on the cell surface. The antibodies associated with a given lymphocyte react to one type of antigen (more precisely to a small number of structurally similar antigens). When the antibodies recognize appropriate antigen they stimulate what results in intensive cloning, and the number of new clones is proportional to the degree of affinity between antibody and the antigen. This process is referred to as clonal selection. It is responsible for maintaining sufficient diversity of B-cells repertoire. To increase defense abilities of the immune system, the clones are subjected somatic mutation, i.e. mutation with very high rate. This way new, well fitted to the intruder, cells are entered to the system. Ineffective mutated clones as well as ineffective B-cells (which for a longer time do not participate in the immune response) are removed from the organism. This process is said to be apoptosis, or programmed death of cells. In place of ineffective cells new, almost randomly produced, cells are entered. Daily about 5% of B-lymphocytes is replaced by newly produced cells. More detailed system activity can be found in [6] or [7]. The process of production new antibodies fitted to an antigen that enters organism for the first time is referred to as primary immune response. It takes time (about three weeks) to produce effective antibodies. When the antigen enters organism one more time, the immune response called secondary immune response is much more efficient. The appropriate antibodies are produced very quickly and in much more amount. The effectivity of the secondary response can be explained by the existence of immune memory. Organism memorizes antigens entering it, and during secondary attack of an antigen, or a pathogen structurally similar to already known intruder, it quickly recalls appropriate antibodies. Interestingly, the nature of immune memory is not precisely known. According to Jerne s hypothesis, [8], B-cells are organized into so-called idiotypic network. Although not confirmed by immunologists this hypothesis offers an interesting and valuable metaphor for constructing systems for data analysis, [9]. The main mechanism responsible for the introduction of new cells and for maintaining efficient network is so-called meta-dynamics which controls the concentration of different kinds of B-cells according to the equation, [1] rate of population diversity = production of new cells death of ineffective cells + reproduction of stimulated cells While clonal selection, somatic mutation and affinity maturation (i.e. maintaining effective B-cells) resemble mechanisms used by the evolutionary algorithms, metadynamics is the unique feature of AISs. As stated by Bersini and Varela, [11], in an ecosystem the species population densities vary according to the interactions with other members of the network as well as through environmental impacts. In addition the whole network is subjected structural perturbations through appearance and disappearance of some species. A crucial feature of AIS is the fact that the network as such, and not the environment, exerts the greatest pressure in the selection of the new cells to be integrated to the network. (1) 3

4 3. Optimization of non-stationary functions The task of non-stationary functions optimization is the identification of a series of optima that change their location (and possibly their height) in time. Since each optimum is located in different point of the search space that is represented by different chromosome, the algorithm designed to cope with this task can be viewed as pattern tracking algorithm. More formally we want to identify all the optima of a function f(x, t) where x D R m and t represents time. Typically the domain D is the Cartesian product of the intervals [x i,min, x i,max ], i = 1,, m. Evolutionary algorithms designed to cope with such stated task exploit one of the following strategies, [12]: the expansion of the memory in order to build up a repertoire of ready responses for environmental changes, or the application of some mechanism for increasing population diversity in order to compensate for changes encountered in the environment. In this last case commonly used mechanisms are: random immigrants mechanism, triggered hypermutation or simply increasing the mutation rate within a standard GA to a constant high level. The first immune algorithm, called Simple Artificial Immune System or Sais, to cope with pattern tracking in dynamic environment was proposed in [3]. Here population consists of B-cells represented as binary strings, that is the Sais is so-called binary immune system. There is no distinction among a B-cell and the antibodies located on its surface (in fact all these antibodies recognize the same antigens). Thus we can use interchangeably the term antibody and B-cell. The algorithm uses mechanisms described in Section 2. Particularly, to measure the affinity of an antibody to currently presented antigen (i.e. optimum at given iteration) so-called exogenic activation is used (defined as the Hamming distance between antigen and an antibody) and the affinity of an antibody to other antibodies is measured in terms of so-called endogenic activation. Later these authors proposed YaSais (Yet another Sais) in which only exogenic activation was taken into account. In this paper we use the algorithm tested already in [5]. It is also a binary immune system and its pseudocode is given in Figure Fitness evaluation. For each individual or antibody p in the population P compute its fitness i.e. the value of the objective function f p. 2. Clonal selection. Choose n antibodies with highest fitness to the antigen. 3. Somatic hypermutation. Make c i mutated clones of i-th antibody. The clone c (i) with highest fitness replaces original antibody if f c(i) > f i. 4. Apoptosis. Each t d 1 iterations, replace d weakest antibodies by randomly generated binary strings. Figure 1. Frame immune algorithm used in the experiments described later 4

5 Step 2 of this algorithm can be realized at two different ways. Choosing antibodies for cloning we can use their genotypic or phenotypic affinity. However it was observed in [5] that genotypic affinity controls the algorithm behavior more efficiently. This affinity is computed as follows. Suppose i* is an individual with highest fitness value. Call this individual current antigen. Now we compare pointwisely (separately on each segment corresponding to different dimension) current antigen with i-th antibody. If the two strings agree on j-th position the affinity is increased by 2 L-j-1. Figure 2 illustrates exemplary computation of the affinity under the assumption that the binary strings representing 2-dimensional real vectors. antigen i*: antibody i: weight aff(i,i*) ( ) + (8+2+1) = 38 Figure 2. Computing the affinity between antigen and an antibody 4. Experiments We performed a set of experiments with the algorithm described in previous section. We did three groups of experiments with two types of environments. Our test-bed was a test-case generator proposed in [13]. The generator creates a convex search space, which is a multidimensional hypercube. It is divided into a number of disjoint subspaces of the same size with defined simple unimodal functions of the same shape but possibly different value of optimum. In case of two-dimensional search space we simply have a patchy landscape, i.e. a chess-board with a hill in the middle of every field. Hills do not move but cyclically change their heights what makes the landscape varying in time. The goal is to find the current highest hill. In our experiments there was a sequence of fields with varying hills heights. Other fields of the space were static. We did experiments with twodimensional search space where the chess-boards were of size 4 by 4, i.e. with 16 fields, and of size 6 by 6, i.e. with 36 fields. Thus the search spaces consisted of 15 local optima and one global optimum in the first case, and of 35 local optima and one global optimum in the second one. We tested four shapes of the sequence of non-stationary fields presented in Figure 3. In the figure, values in cells are weights of unimodal fuctions of the respective fields, which control heights of the hills. In other words the function located at the (i,j)-field is of the form f ij (x,y) = w ij g(x-a i, y-b j ), where g is a fixed unimodal function and (a i, b j ) is the center of this field. Lower index at the value in the cell represents the position of the field in the sequence of presented optima. The environments #1 and #3 (left part of Figure 3) were test-beds for experiments with cyclic changes, while the environments #2 and #4 (right part of Figure 3) were test-beds for experiments with both cyclic and acyclic changes. The aim of these experiments was to trace efficiency of primary (acyclic changes) and secondary (cyclic changes) immune response to the antigens (i.e. current 5

6 optima). For experiments with cyclic changes a single epoch obeys 5 cycles of changes. In all the experiments each antigen has been presented through 1 iterations. Thus, in case of the environment #1 a single epoch took 2 iterations, in case of the environment #2-4 iterations, in case of the environment #3-3 iterations, and in case of the environment #4-6 iterations. Experiments with non-cyclic changes were based on the environments #2 and #4 and a single epoch included just one cycle of changes and took 8 and 12 iterations respectively Figure 3. Environments #1 (top left), #2 (top right), #3 (bottom left) and #4 (bottom right) - shapes of the sequence of non-stationary fields in testing environments. For the six environments described above we did series of experiments by changing the parameters n (step 2 of the algorithm in Figure 1) see Figure 4. % 8% 6% 4% exogenic activation immune memory apoptosis 2% % Figure 4. Division of population into activated individuals and individuals for apoptosis. Individuals that do not belong to any of the two subgroups are supposed to be an immune memory structure. 6

7 Every experiment of the seven from the Figure 4 was repeated through 5 epochs and in the later figures we always study average values of these 5 epochs. For the results estimation we used two measures proposed in [13]: Accuracy and Adaptability. Accuracy is a difference between the value of the current best individual in the population of the just before the change generation and the optimum value averaged over the entire run. Adaptability is a difference between the value of the current best individual of each generation and the optimum value averaged over the entire run. For both measures the smaller values are the better results. 4.1 Cyclic changes The results of cyclic changes are presented in Figure 5 (env. #1) Figure 6 (env. #2), Figure 6 (env. #3), and Figure 6 (env. #4) avg value for 5 experiments Acc Ada Figure 5. Results for experiments with 7 types of parameter settings performed with environment #1 (cyclic changes). avg value for 5 experiments Acc Ada Figure 6. Results for experiments with 7 types of parameter settings performed with environment #2 (cyclic changes). 7

8 avg value for 5 experiments Acc Ada Figure 7. Results for experiments with 7 types of parameter settings performed with environment #3 (cyclic changes). avg value for 5 experiments Acc Ada Figure 8. Results for experiments with 7 types of parameter settings performed with environment #4 (cyclic changes). In the figures the best results, i.e. the smallest values of Accuracy and Adaptability are obtained for different. However, every time they are better than for the algorithm where all individuals are activated or undergo apoptosis (settings No. 1) and better than the case where the number of activated individuals is small. For the environment #1 the best case is setting No. 5, the environment #2 No. 3, the environments #3 and #4 No Non-cyclic changes The results of non-cyclic changes are presented in Figure 9 and Figure 1. In this case the best results are obtained for the settings where all individuals are activated or undergo apoptosis (setting No. 1) and the worst are in the case where the number of activated individuals is the smallest. 8

9 avg value for 5 experiments Acc Ada Figure 9. Results for experiments with 7 types of parameter settings performed with environment #2 (non-cyclic changes). avg value for 5 experiments Acc Ada Figure 1. Results for experiments with 7 types of parameter settings performed with environment #4 (non-cyclic changes). 5. Conclusions Obtained results confirmed, that the individuals, which belong neither to activated group nor to the group for apoptosis play significant role in searching for optima in problems with cyclic changes only. Otherwise, when the problem changes in any non-cyclic way, the best approach is to divide the population simply into two groups: an activated individuals group and a group of individuals for apoptosis. For environments with cyclic changes we can not propose an effective general proportion between the groups in a population. Differences between the best settings of algorithm parameters indicate, that the best proportion depends of the type of changes in optimised environment and should be tuned individually. 9

10 But anyway, we can ascertain, that in case of cyclic changes a key to a success is a balance between explorative mechanisms of the algorithm (represented by phases of activation apoptosis) and from the other side - a form of immune memory represented by the left group of individuals, that does not take part in both phases. Bibliography [1] Nowell, A., Simon, H.A. Human Problem Solving. Prentice Hall, NJ 1972 [2] Holland, J.H. Adaptation in Natural and Artificial Systems. MIT Press 1992 [3] Gaspar, A., Collard, Ph. From Gas to artificial immune systems: Improving adaptation in time dependent optimization. Proc. of the 1999 Congress on Evolutionary Computation, [4] Wierzchoń, S.T. Multimodal optimization with artificial immune systems. M.A. Kłopotek, M.Michalewicz, S.T.Wierzchoń, eds, Intelligent Information Systems 21. Physica-Verlag 21, [5] Wierzchoń, S.T. Artificial immune systems in action: Optimization of nonstationary functions (in Polish). Proc. of the Workshop Artificial Intelligence, SzI 21, Siedlce, Poland, December [6] Hofmeyr, S.A. An interpretative introduction to the immune system. Technical Report, Dept. of Computer Science, University of New Mexico, Albuquerque, NM, 1999 [7] Wierzchoń, S.T. Artificial Immune Systems. Theory and Applications (in Polish). Warszawa 21 [8] Jerne, N.J. Towards a network theory of the immune system. Ann. Immunol. (Inst. Pasteur), 125C: , 1974 [9] de Castro, L.N., von Zuben, F.J. ainet: An artificial immune network for data analysis. In: H.A. Abbas, R. A. Sarker, Ch. S. Newton (eds.) Data Mining: A Heuristic Approach. Idea Group Publishing, USA, 21 [1] Perelson, A.S. Immune network theory. Immunological Review, 11: 5-36, 1989 [11] Bersini,H., Varela, F. The immune learning mechanisms: Reinforcement and recruitment and their applications. Computing with Biological Metaphors, Chapman Hall, 1994, [12] Cobb, H.G., Grefenstette, J.J. Genetic algorithms for tracking changing environments. In: Proceedings of the Fifth International Conference on Genetic Algorithms. Morgan Kaufmann 1993 [13] Trojanowski,K., and Michalewicz, Z., Searching for optima in nonstationary environments, CEC 99, IEEE Publishing, pp

Stable Clusters Formation in an Artificial Immune System

Stable Clusters Formation in an Artificial Immune System Stable Clusters Formation in an Artificial Immune System S.T. Wierzchoń Department of Computer Science, Białystok Technical University ul. Wiejska 45 a, 15-351 Białystok, Poland and Institute of Computer

More information

FUNCTION OPTIMIZATION BY THE IMMUNE METAPHOR

FUNCTION OPTIMIZATION BY THE IMMUNE METAPHOR TASK QUARTERLY 6 No 3 (2002), 1 16 FUNCTION OPTIMIZATION BY THE IMMUNE METAPHOR SŁAWOMIR T. WIERZCHOŃ Institute o Computer Science, Polish Academy of Sciences, Ordona 21, 01-267 Warsaw, Poland stw@ipipan.waw.pl

More information

Artificial Immune Systems Tutorial

Artificial Immune Systems Tutorial Artificial Immune Systems Tutorial By Dr Uwe Aickelin http://www.aickelin.com Overview Biological Immune System. Artificial Immune System (AIS). Comparison to other Algorithms. Applications of AIS: Data

More information

The Nottingham eprints service makes this work by researchers of the University of Nottingham available open access under the following conditions.

The Nottingham eprints service makes this work by researchers of the University of Nottingham available open access under the following conditions. Aickelin, Uwe (2003) Artificial Immune System and Intrusion Detection Tutorial. In: Introduction Tutorials in Optimization, Search and Decision Support Methodologies, Nottingham, UK. Access from the University

More information

Artificial Immune Systems

Artificial Immune Systems Artificial Immune Systems Dr. Mario Pavone Department of Mathematics & Computer Science University of Catania mpavone@dmi.unict.it http://www.dmi.unict.it/mpavone/ Biological Immune System (1/4) Immunology

More information

Metodi e tecniche di ottimizzazione innovative per applicazioni elettromagnetiche

Metodi e tecniche di ottimizzazione innovative per applicazioni elettromagnetiche Metodi e tecniche di ottimizzazione innovative per applicazioni elettromagnetiche Algoritmi stocastici Parte 3 Artificial Immune Systems M. Repetto Dipartimento Ingegneria Elettrica Industriale - Politecnico

More information

Solving Protein Folding Problem Using Hybrid Genetic Clonal Selection Algorithm

Solving Protein Folding Problem Using Hybrid Genetic Clonal Selection Algorithm 94 Solving Protein Folding Problem Using Hybrid Genetic Clonal Selection Algorithm Adel Omar Mohamed and Abdelfatah A. Hegazy, Amr Badr College of Computing & Information Technology, Arab Academy Abstract:

More information

Immune Programming. Payman Samadi. Supervisor: Dr. Majid Ahmadi. March Department of Electrical & Computer Engineering University of Windsor

Immune Programming. Payman Samadi. Supervisor: Dr. Majid Ahmadi. March Department of Electrical & Computer Engineering University of Windsor Immune Programming Payman Samadi Supervisor: Dr. Majid Ahmadi March 2006 Department of Electrical & Computer Engineering University of Windsor OUTLINE Introduction Biological Immune System Artificial Immune

More information

Improved Clonal Selection Algorithm (ICLONALG)

Improved Clonal Selection Algorithm (ICLONALG) International Journal of Current Engineering and Technology E-ISSN 2277 4106, P-ISSN 2347 5161 2015INPRESSCO, All Rights Reserved Available at http://inpressco.com/category/ijcet Research Article Nidhi

More information

Clonal Selection Method for Virus Detection in a Cloud

Clonal Selection Method for Virus Detection in a Cloud Clonal Selection Method for Virus Detection in a Cloud Agnika Sahu #1, Tanmaya Swain *2, Tapaswini Samant *3 # School of Computer Engineering, KIIT University Bhubaneswar, India Abstract The biological

More information

An Introduction to Artificial Immune Systems

An Introduction to Artificial Immune Systems An Introduction to Artificial Immune Systems Jonathan Timmis Computing Laboratory University of Kent at Canterbury CT2 7NF. UK. J.Timmis@kent.ac.uk http:/www.cs.kent.ac.uk/~jt6 AIS October 2003 1 Novel

More information

ARTIFICIAL IMMUNE SYSTEM: ALGORITHMS AND APPLICATIONS REVIEW

ARTIFICIAL IMMUNE SYSTEM: ALGORITHMS AND APPLICATIONS REVIEW ARTIFICIAL IMMUNE SYSTEM: ALGORITHMS AND APPLICATIONS REVIEW Pankaj Chaudhary Student at IMS Engineering College, Ghaziabad pchaudhary929@gmail.com Kundan Kumar Student at IMS engineering college, Ghaziabad

More information

An Overview of Artificial Immune Systems

An Overview of Artificial Immune Systems An Overview of Artificial Immune Systems J. Timmis 1*, T. Knight 1, L.N. de Castro 2 and E. Hart 3 1 Computing Laboratory, University of Kent. Canterbury. UK. {jt6,tpk1}@ukc.ac.uk 2 School of Electrical

More information

MINIMIZE THE MAKESPAN FOR JOB SHOP SCHEDULING PROBLEM USING ARTIFICIAL IMMUNE SYSTEM APPROACH

MINIMIZE THE MAKESPAN FOR JOB SHOP SCHEDULING PROBLEM USING ARTIFICIAL IMMUNE SYSTEM APPROACH MINIMIZE THE MAKESPAN FOR JOB SHOP SCHEDULING PROBLEM USING ARTIFICIAL IMMUNE SYSTEM APPROACH AHMAD SHAHRIZAL MUHAMAD, 1 SAFAAI DERIS, 2 ZALMIYAH ZAKARIA 1 Professor, Faculty of Computing, Universiti Teknologi

More information

Pattern Recognition Approaches inspired by Artificial Immune System

Pattern Recognition Approaches inspired by Artificial Immune System Pattern Recognition Approaches inspired by Artificial Immune System Aanchal Malhotra Technology, Amity University, Abhishek Baheti Technology, Amity University, Shilpi Gupta Technology, Amity University

More information

A Comparative Study of Immune System Based Genetic Algorithms in Dynamic Environments

A Comparative Study of Immune System Based Genetic Algorithms in Dynamic Environments A Comparative Study of Immune System Based Genetic Algorithms in Dynamic Environments Shengxiang Yang Department of Computer Science, University of Leicester University Road, Leicester LE1 7RH, United

More information

A Gene Based Adaptive Mutation Strategy for Genetic Algorithms

A Gene Based Adaptive Mutation Strategy for Genetic Algorithms A Gene Based Adaptive Mutation Strategy for Genetic Algorithms Sima Uyar, Sanem Sariel, and Gulsen Eryigit Istanbul Technical University, Electrical and Electronics Faculty Department of Computer Engineering,

More information

Chapter 4. Artificial Immune Systems

Chapter 4. Artificial Immune Systems Chapter 4 Artificial Immune Systems The different theories in the science of immunology inspired the development (design) of immune inspired algorithms, collectively known as artificial immune systems

More information

SEISMIC ATTRIBUTES SELECTION AND POROSITY PREDICTION USING MODIFIED ARTIFICIAL IMMUNE NETWORK ALGORITHM

SEISMIC ATTRIBUTES SELECTION AND POROSITY PREDICTION USING MODIFIED ARTIFICIAL IMMUNE NETWORK ALGORITHM Journal of Engineering Science and Technology Vol. 13, No. 3 (2018) 755-765 School of Engineering, Taylor s University SEISMIC ATTRIBUTES SELECTION AND POROSITY PREDICTION USING MODIFIED ARTIFICIAL IMMUNE

More information

Implementation of Artificial Immune System Algorithms

Implementation of Artificial Immune System Algorithms Implementation of Artificial Immune System Algorithms K. Sri Lakshmi Associate Professor, Department of CSE Abstract An artificial immune system (AIS) that is distributed, robust, dynamic, diverse and

More information

An optimization framework for modeling and simulation of dynamic systems based on AIS

An optimization framework for modeling and simulation of dynamic systems based on AIS Title An optimization framework for modeling and simulation of dynamic systems based on AIS Author(s) Leung, CSK; Lau, HYK Citation The 18th IFAC World Congress (IFAC 2011), Milano, Italy, 28 August-2

More information

ARTIFICIAL IMMUNE ALGORITHMS IN LEARNING AND OPTIMIZATION

ARTIFICIAL IMMUNE ALGORITHMS IN LEARNING AND OPTIMIZATION ARTIFICIAL IMMUNE ALGORITHMS IN LEARNING AND OPTIMIZATION and Kevin Sim Edinburgh Napier University, Scotland, UK Keywords: Artificial Immune Systems, immunology, optimization, classification, clustering,

More information

IMPLEMENTATION OF AN OPTIMIZATION TECHNIQUE: GENETIC ALGORITHM

IMPLEMENTATION OF AN OPTIMIZATION TECHNIQUE: GENETIC ALGORITHM IMPLEMENTATION OF AN OPTIMIZATION TECHNIQUE: GENETIC ALGORITHM TWINKLE GUPTA* Department of Computer Science, Hindu Kanya MahaVidyalya, Jind, India Abstract We are encountered with various optimization

More information

A Scalable Artificial Immune System Model for Dynamic Unsupervised Learning

A Scalable Artificial Immune System Model for Dynamic Unsupervised Learning A Scalable Artificial Immune System Model for Dynamic Unsupervised Learning Olfa Nasraoui, Fabio Gonzalez 2, Cesar Cardona, Carlos Rojas, and Dipankar Dasgupta 2 Department of Electrical and Computer Engineering,

More information

EVOLUTIONARY ALGORITHMS AT CHOICE: FROM GA TO GP EVOLŪCIJAS ALGORITMI PĒC IZVĒLES: NO GA UZ GP

EVOLUTIONARY ALGORITHMS AT CHOICE: FROM GA TO GP EVOLŪCIJAS ALGORITMI PĒC IZVĒLES: NO GA UZ GP ISSN 1691-5402 ISBN 978-9984-44-028-6 Environment. Technology. Resources Proceedings of the 7 th International Scientific and Practical Conference. Volume I1 Rēzeknes Augstskola, Rēzekne, RA Izdevniecība,

More information

A Fractal Immune Network

A Fractal Immune Network A Fractal Immune Network Peter J. Bentley 1 and Jon Timmis 2 1 Department of Computer Science, University College London. UK p.bentley@cs.ucl.ac.uk http://www.cs.ucl.ac.uk/staff/p.bentley/ 2 Computing

More information

Biological immune systems

Biological immune systems Immune Systems 1 Introduction 2 Biological immune systems Living organism must protect themselves from the attempt of other organisms to exploit their resources Some would-be exploiter (pathogen) is much

More information

BIPOLAR CONVERGENCE IN GENETIC ALGORITHM FOR MULTIMODAL OPTIMIZATION

BIPOLAR CONVERGENCE IN GENETIC ALGORITHM FOR MULTIMODAL OPTIMIZATION Proceedings of the International Conference on Theory and Applications of Mathematics and Informatics ICTAMI 2003, Alba Iulia BIPOLAR CONVERGENCE IN GENETIC ALGORITHM FOR MULTIMODAL OPTIMIZATION by Corina

More information

Genetic algorithms. History

Genetic algorithms. History Genetic algorithms History Idea of evolutionary computing was introduced in the 1960s by I. Rechenberg in his work "Evolution strategies" (Evolutionsstrategie in original). His idea was then developed

More information

An Efficient and Effective Immune Based Classifier

An Efficient and Effective Immune Based Classifier Journal of Computer Science 7 (2): 148-153, 2011 ISSN 1549-3636 2011 Science Publications An Efficient and Effective Immune Based Classifier Shahram Golzari, Shyamala Doraisamy, Md Nasir Sulaiman and Nur

More information

Introduction to Artificial Intelligence. Prof. Inkyu Moon Dept. of Robotics Engineering, DGIST

Introduction to Artificial Intelligence. Prof. Inkyu Moon Dept. of Robotics Engineering, DGIST Introduction to Artificial Intelligence Prof. Inkyu Moon Dept. of Robotics Engineering, DGIST Chapter 9 Evolutionary Computation Introduction Intelligence can be defined as the capability of a system to

More information

AN ALGORITHM FOR REMOTE SENSING IMAGE CLASSIFICATION BASED ON ARTIFICIAL IMMUNE B CELL NETWORK

AN ALGORITHM FOR REMOTE SENSING IMAGE CLASSIFICATION BASED ON ARTIFICIAL IMMUNE B CELL NETWORK AN ALGORITHM FOR REMOTE SENSING IMAGE CLASSIFICATION BASED ON ARTIFICIAL IMMUNE B CELL NETWORK Shizhen Xu a, *, Yundong Wu b, c a Insitute of Surveying and Mapping, Information Engineering University 66

More information

CHAPTER 4 PROPOSED HYBRID INTELLIGENT APPROCH FOR MULTIPROCESSOR SCHEDULING

CHAPTER 4 PROPOSED HYBRID INTELLIGENT APPROCH FOR MULTIPROCESSOR SCHEDULING 79 CHAPTER 4 PROPOSED HYBRID INTELLIGENT APPROCH FOR MULTIPROCESSOR SCHEDULING The present chapter proposes a hybrid intelligent approach (IPSO-AIS) using Improved Particle Swarm Optimization (IPSO) with

More information

The 'Pathogenic Exposure' Paradigm

The 'Pathogenic Exposure' Paradigm The 'Pathogenic Exposure' Paradigm JASON BROWNLEE Technical Report 070422A Complex Intelligent Systems Laboratory, Centre for Information Technology Research, Faculty of Information and Communication Technologies,

More information

DEVELOPMENT OF MULTI-OBJECTIVE SIMULATION-BASED GENETIC ALGORITHM FOR SUPPLY CHAIN CYCLIC PLANNING AND OPTIMISATION

DEVELOPMENT OF MULTI-OBJECTIVE SIMULATION-BASED GENETIC ALGORITHM FOR SUPPLY CHAIN CYCLIC PLANNING AND OPTIMISATION From the SelectedWorks of Liana Napalkova May, 2008 DEVELOPMENT OF MULTI-OBJECTIVE SIMULATION-BASED GENETIC ALGORITHM FOR SUPPLY CHAIN CYCLIC PLANNING AND OPTIMISATION Galina Merkuryeva Liana Napalkova

More information

Modelling as a way in design of novel algorithms in computational intelligence

Modelling as a way in design of novel algorithms in computational intelligence Modelling as a way in design of novel algorithms in computational intelligence Helena Szczerbicka Modelling and Simulation Group Computer Science & Electrical Engineering Faculty Leibniz University of

More information

Kent Academic Repository

Kent Academic Repository Kent Academic Repository Full text document (pdf) Citation for published version de Castro, Leandro N. and Timmis, Jon (2002) Artificial Immune Systems: A Novel Approach to Pattern Recognition. In: Corchado,

More information

Artificial Immune Systems: Using the Immune System as Inspiration for Data Mining

Artificial Immune Systems: Using the Immune System as Inspiration for Data Mining Artificial Immune Systems 209 Chapter XI Artificial Immune Systems: Using the Immune System as Inspiration for Data Mining Jon Timmis and Thomas Knight University of Kent at Canterbury, UK The immune system

More information

TIMETABLING EXPERIMENTS USING GENETIC ALGORITHMS. Liviu Lalescu, Costin Badica

TIMETABLING EXPERIMENTS USING GENETIC ALGORITHMS. Liviu Lalescu, Costin Badica TIMETABLING EXPERIMENTS USING GENETIC ALGORITHMS Liviu Lalescu, Costin Badica University of Craiova, Faculty of Control, Computers and Electronics Software Engineering Department, str.tehnicii, 5, Craiova,

More information

The Human Immune System and Network Intrusion Detection

The Human Immune System and Network Intrusion Detection The Human Immune System and Network Intrusion Detection Jungwon Kim and Peter Bentley Department of Computer Science, University Collge London Gower Street, London, WC1E 6BT, U. K. Phone: +44-171-380-7329,

More information

Optimisation and Operations Research

Optimisation and Operations Research Optimisation and Operations Research Lecture 17: Genetic Algorithms and Evolutionary Computing Matthew Roughan http://www.maths.adelaide.edu.au/matthew.roughan/ Lecture_notes/OORII/

More information

Artificial Immune-Based For Voltage Stability Prediction In Power System

Artificial Immune-Based For Voltage Stability Prediction In Power System Artificial Immune-Based For Voltage Stability Prediction In Power System S. I. Suliman, T. K. Abdul Rahman, I. Musirin Faculty of Electrical Engineering, Universiti Teknologi MARA,40450, Shah Alam, Selangor

More information

ARTICLE IN PRESS. Immune programming

ARTICLE IN PRESS. Immune programming Information Sciences xxx (2005) xxx xxx www.elsevier.com/locate/ins Immune programming Petr Musilek *, Adriel Lau, Marek Reformat, Loren Wyard-Scott Department of Electrical and Computer Engineering, W2-030

More information

PDGA: the Primal-Dual Genetic Algorithm

PDGA: the Primal-Dual Genetic Algorithm P: the Primal-Dual Genetic Algorithm Shengxiang Yang Department of Computer Science University of Leicester University Road, Leicester LE1 7RH, UK Email: syang@mcsleacuk Abstract Genetic algorithms (GAs)

More information

STRUCTURAL OPTIMIZATION USING ARTIFICIAL IMMUNE SYSTEM (AIS)

STRUCTURAL OPTIMIZATION USING ARTIFICIAL IMMUNE SYSTEM (AIS) Blucher Mechanical Engineering Proceedings May 2014, vol. 1, num. 1 www.proceedings.blucher.com.br/evento/10wccm STRUCTURAL OPTIMIZATION USING ARTIFICIAL IMMUNE SYSTEM (AIS) Sai Sushank Botu 1, S V Barai

More information

ainet: An Artificial Immune Network for Data Analysis

ainet: An Artificial Immune Network for Data Analysis ainet: An Artificial Immune Network for Data Analysis Leandro Nunes de Castro & Fernando José Von Zuben {lnunes,vonzuben}@dca.fee.unicamp.br http://www.dca.fee.unicamp.br/~lnunes ftp://ftp.dca.fee.unicamp.br/pub/docs/vonzuben/lnunes/dmha.pdf

More information

Journal of Global Research in Computer Science PREMATURE CONVERGENCE AND GENETIC ALGORITHM UNDER OPERATING SYSTEM PROCESS SCHEDULING PROBLEM

Journal of Global Research in Computer Science PREMATURE CONVERGENCE AND GENETIC ALGORITHM UNDER OPERATING SYSTEM PROCESS SCHEDULING PROBLEM Volume, No. 5, December 00 Journal of Global Research in Computer Science RESEARCH PAPER Available Online at www.jgrcs.info PREMATURE CONVERGENCE AND GENETIC ALGORITHM UNDER OPERATING SYSTEM PROCESS SCHEDULING

More information

Fixed vs. Self-Adaptive Crossover-First Differential Evolution

Fixed vs. Self-Adaptive Crossover-First Differential Evolution Applied Mathematical Sciences, Vol. 10, 2016, no. 32, 1603-1610 HIKARI Ltd, www.m-hikari.com http://dx.doi.org/10.12988/ams.2016.6377 Fixed vs. Self-Adaptive Crossover-First Differential Evolution Jason

More information

An Investigation of a Methodology for the Development of Artificial Immune Systems: A Case-Study in Immune Receptor Degeneracy

An Investigation of a Methodology for the Development of Artificial Immune Systems: A Case-Study in Immune Receptor Degeneracy An Investigation of a Methodology for the Development of Artificial Immune Systems: A Case-Study in Immune Receptor Degeneracy Paul Simon Andrews Submitted for the degree of Doctor of Philosophy University

More information

A Resource Limited Artificial Immune System for Data. Analysis. Jon Timmis* and Mark Neal **

A Resource Limited Artificial Immune System for Data. Analysis. Jon Timmis* and Mark Neal ** A Resource Limited Artificial Immune System for Data Analysis Jon Timmis* and Mark Neal ** * Computing Laboratory **Department of Computer Science University of Kent at Canterbury Canterbury, Kent. UK.

More information

What is Evolutionary Computation? Genetic Algorithms. Components of Evolutionary Computing. The Argument. When changes occur...

What is Evolutionary Computation? Genetic Algorithms. Components of Evolutionary Computing. The Argument. When changes occur... What is Evolutionary Computation? Genetic Algorithms Russell & Norvig, Cha. 4.3 An abstraction from the theory of biological evolution that is used to create optimization procedures or methodologies, usually

More information

Deterministic Crowding, Recombination And Self-Similarity

Deterministic Crowding, Recombination And Self-Similarity Deterministic Crowding, Recombination And Self-Similarity Bo Yuan School of Information Technology and Electrical Engineering The University of Queensland Brisbane, Queensland 4072 Australia E-mail: s4002283@student.uq.edu.au

More information

Evolutionary Developmental System for Structural Design

Evolutionary Developmental System for Structural Design Evolutionary Developmental System for Structural Design Rafal Kicinger George Mason University 4400 University Drive MS 4A6 Fairfax, VA 22030 rkicinge@gmu.edu Abstract This paper discusses the results

More information

GENETIC ALGORITHMS. Narra Priyanka. K.Naga Sowjanya. Vasavi College of Engineering. Ibrahimbahg,Hyderabad.

GENETIC ALGORITHMS. Narra Priyanka. K.Naga Sowjanya. Vasavi College of Engineering. Ibrahimbahg,Hyderabad. GENETIC ALGORITHMS Narra Priyanka K.Naga Sowjanya Vasavi College of Engineering. Ibrahimbahg,Hyderabad mynameissowji@yahoo.com priyankanarra@yahoo.com Abstract Genetic algorithms are a part of evolutionary

More information

ARTIFICIAL IMMUNE SYSTEM AGENT MODEL

ARTIFICIAL IMMUNE SYSTEM AGENT MODEL ARTIFICIAL IMMUNE SYSTEM AGENT MODEL Siti Mazura Che Doi 1 and Norita Md. Norwawi 2 Universiti Sains Islam Malaysia (USIM) {sitimazura@ipip.edu.my, norita}@usim.edu.my ABSTRACT. The Artificial Systems

More information

initial set of random solutions called population satisfying boundary and/or system

initial set of random solutions called population satisfying boundary and/or system CHAPTER 4 Genetic Algorithm GAs are stochastic search algorithms based on the mechanism of natural selection and natural genetics. GA, differing from conventional search techniques, start with an initial

More information

From Genetics to Genetic Algorithms

From Genetics to Genetic Algorithms From Genetics to Genetic Algorithms Solution to Optimisation Problems Using Natural Systems Jitendra R Raol and Abhijit Jalisatgi Genetic algorithms are search procedures inspired by natural selection

More information

Data Selection for Semi-Supervised Learning

Data Selection for Semi-Supervised Learning Data Selection for Semi-Supervised Learning Shafigh Parsazad 1, Ehsan Saboori 2 and Amin Allahyar 3 1 Department Of Computer Engineering, Ferdowsi University of Mashhad Mashhad, Iran Shafigh.Parsazad@stu-mail.um.ac.ir

More information

CHAPTER 7 CELLULAR BASIS OF ANTIBODY DIVERSITY: CLONAL SELECTION

CHAPTER 7 CELLULAR BASIS OF ANTIBODY DIVERSITY: CLONAL SELECTION CHAPTER 7 CELLULAR BASIS OF ANTIBODY DIVERSITY: CLONAL SELECTION The specificity of humoral immune responses relies on the huge DIVERSITY of antigen combining sites present in antibodies, diversity which

More information

Artificial Immune Systems: Theory and Applications

Artificial Immune Systems: Theory and Applications Artificial Immune Systems: Theory and Applications Leandro Nunes de Castro Financial Support: FAPESP 98/11333-9 lnunes@dca.fee.unicamp.br State University of Campinas - UNICAMP School of Computer and Electrical

More information

Artificial Immune System

Artificial Immune System Artificial Immune System 1 Introduction Magnus Erik Hvass Pedersen (971055) Daimi, University of Aarhus, May 2003 The purpose of this document is to verify attendance of the author to the Swarm Intelligence

More information

Research on Intrusion Detection based on Immunology Principle. Guannan GONG

Research on Intrusion Detection based on Immunology Principle. Guannan GONG Research on Intrusion Detection based on Immunology Principle Guannan GONG beckhangong@hotmail.com Liang HU hul@mail.jlu.edu.cn College of Computer Science and Technology, Jilin University, Changchun,

More information

Rule Minimization in Predicting the Preterm Birth Classification using Competitive Co Evolution

Rule Minimization in Predicting the Preterm Birth Classification using Competitive Co Evolution Indian Journal of Science and Technology, Vol 9(10), DOI: 10.17485/ijst/2016/v9i10/88902, March 2016 ISSN (Print) : 0974-6846 ISSN (Online) : 0974-5645 Rule Minimization in Predicting the Preterm Birth

More information

An introduction to evolutionary computation

An introduction to evolutionary computation An introduction to evolutionary computation Andrea Roli andrea.roli@unibo.it Dept. of Computer Science and Engineering (DISI) Campus of Cesena Alma Mater Studiorum Università di Bologna Outline 1 Basic

More information

ARTIFICIAL IMMUNE SYSTEM CLASSIFICATION OF MULTIPLE- CLASS PROBLEMS

ARTIFICIAL IMMUNE SYSTEM CLASSIFICATION OF MULTIPLE- CLASS PROBLEMS 1 ARTIFICIAL IMMUNE SYSTEM CLASSIFICATION OF MULTIPLE- CLASS PROBLEMS DONALD E. GOODMAN, JR. Mississippi State University Department of Psychology Mississippi State, Mississippi LOIS C. BOGGESS Mississippi

More information

Evolutionary algorithms to simulate the phylogenesis of a binary artificial immune system

Evolutionary algorithms to simulate the phylogenesis of a binary artificial immune system Evol. Intel. (2008) 1:133 144 DOI 10.1007/s12065-008-0010-z RESEARCH PAPER Evolutionary algorithms to simulate the phylogenesis of a binary artificial immune system Grazziela P. Figueredo Æ Luis A. V.

More information

GENETIC ALGORITHM BASED APPROACH FOR THE SELECTION OF PROJECTS IN PUBLIC R&D INSTITUTIONS

GENETIC ALGORITHM BASED APPROACH FOR THE SELECTION OF PROJECTS IN PUBLIC R&D INSTITUTIONS GENETIC ALGORITHM BASED APPROACH FOR THE SELECTION OF PROJECTS IN PUBLIC R&D INSTITUTIONS SANJAY S, PRADEEP S, MANIKANTA V, KUMARA S.S, HARSHA P Department of Human Resource Development CSIR-Central Food

More information

A Danger-Based Approach to Intrusion Detection

A Danger-Based Approach to Intrusion Detection A Danger-Based Approach to Intrusion Detection Mahdi Zamani *, Mahnush Movahedi *, Mohammad Ebadzadeh, and Hossein Pedram Department of Computer Science, University of New Mexico, Albuquerque, NM, USA

More information

Tuning of 2-DOF PID Controller By Immune Algorithm

Tuning of 2-DOF PID Controller By Immune Algorithm Tuning of 2-DOF PD Controller By mmune Algorithm Dong Hwa Kim Dept. of nstrumentation and Control Eng., Hanbat National University, 16-1 San Duckmyong-Dong Yusong-Gu, Daejon City Seoul, Korea, 305-719.

More information

IMMUNE NETWORK ALGORITHM IN MONTHLY STREAMFLOW PREDICTION AT JOHOR RIVER

IMMUNE NETWORK ALGORITHM IN MONTHLY STREAMFLOW PREDICTION AT JOHOR RIVER IMMUNE NETWORK ALGORITHM IN MONTHLY STREAMFLOW PREDICTION AT JOHOR RIVER Nur Izzah Mat Ali 1, M. A. Malek 2, Amelia Ritahani Ismail 3 1 Department of Civil Engineering, Universiti Tenaga Nasional, Kajang,

More information

Part 1: Motivation, Basic Concepts, Algorithms

Part 1: Motivation, Basic Concepts, Algorithms Part 1: Motivation, Basic Concepts, Algorithms 1 Review of Biological Evolution Evolution is a long time scale process that changes a population of an organism by generating better offspring through reproduction.

More information

Merging Event Logs for Process Mining with Hybrid Artificial Immune Algorithm

Merging Event Logs for Process Mining with Hybrid Artificial Immune Algorithm 10 Int'l Conf. Data Mining DMI'16 Merging Event Logs for Process Mining with Hybrid Artificial Immune Algorithm Yang Xu 1, Qi Lin 1, Martin Q. Zhao 2 1 School of Software Engineering, South China University

More information

Genetic Algorithms and Genetic Programming. Lecture 1: Introduction (25/9/09)

Genetic Algorithms and Genetic Programming. Lecture 1: Introduction (25/9/09) Genetic Algorithms and Genetic Programming Michael Herrmann Lecture 1: Introduction (25/9/09) michael.herrmann@ed.ac.uk, phone: 0131 6 517177, Informatics Forum 1.42 Problem Solving at Decreasing Domain

More information

Immune and Evolutionary Approaches to Software Mutation Testing

Immune and Evolutionary Approaches to Software Mutation Testing Immune and Evolutionary Approaches to Software Mutation Testing Pete May 1, Jon Timmis 2, and Keith Mander 1 1 Computing Laboratory, University of Kent, Canterbury, Kent, UK petesmay@gmail.com, k.c.mander@kent.ac.uk

More information

Genetic Algorithm for Variable Selection. Genetic Algorithms Step by Step. Genetic Algorithm (Holland) Flowchart of GA

Genetic Algorithm for Variable Selection. Genetic Algorithms Step by Step. Genetic Algorithm (Holland) Flowchart of GA http://www.spectroscopynow.com http://ib-poland.virtualave.net/ee/genetic1/3geneticalgorithms.htm http://www.uni-mainz.de/~frosc000/fbg_po3.html relative intensity Genetic Algorithm for Variable Selection

More information

Evolutionary Developmental System for Structural Design 1

Evolutionary Developmental System for Structural Design 1 Evolutionary Developmental System for Structural Design 1 Rafal Kicinger George Mason University 4400 University Drive MS 4A6 Fairfax, VA 22030 rkicinge@gmu.edu Abstract This paper discusses the results

More information

Genetic Algorithm and Neural Network

Genetic Algorithm and Neural Network Proceedings of the 7th WSEAS International Conference on Applied Informatics and Communications, Athens, Greece, August 24-26, 2007 345 Genetic Algorithm and Neural Network JIRI STASTNY*, VLADISLAV SKORPIL**

More information

Immune Network based Ensembles

Immune Network based Ensembles Immune Network based Ensembles Nicolás García-Pedrajas 1 and Colin Fyfe 2 1- Dept. of Computing and Numerical Analysis University of Córdoba (SPAIN) e-mail: npedrajas@uco.es 2- the Dept. of Computing University

More information

A New Approach to Solve Multiple Traveling Salesmen Problem by Clonal Selection Algorithm

A New Approach to Solve Multiple Traveling Salesmen Problem by Clonal Selection Algorithm International Journal of Applied Engineering Research ISSN 0973-4562 Volume 9, Number 21 (2014) pp. 11005-11017 Research India Publications http://www.ripublication.com A New Approach to Solve Multiple

More information

An Artificial Immune System Approach for Flexible Job Shop Scheduling Problem

An Artificial Immune System Approach for Flexible Job Shop Scheduling Problem Proceedings of the 2012 International Conference on Industrial Engineering and Operations Management Istanbul, Turkey, July 3 6, 2012 An Artificial Immune System Approach for Flexible Job Shop Scheduling

More information

A Comparison between Genetic Algorithms and Evolutionary Programming based on Cutting Stock Problem

A Comparison between Genetic Algorithms and Evolutionary Programming based on Cutting Stock Problem Engineering Letters, 14:1, EL_14_1_14 (Advance online publication: 12 February 2007) A Comparison between Genetic Algorithms and Evolutionary Programming based on Cutting Stock Problem Raymond Chiong,

More information

A Course on Meta-Heuristic Search Methods for Combinatorial Optimization Problems

A Course on Meta-Heuristic Search Methods for Combinatorial Optimization Problems A Course on Meta-Heuristic Search Methods for Combinatorial Optimization Problems AutOrI LAB, DIA, Roma Tre Email: mandal@dia.uniroma3.it January 20, 2014 Outline 1 2 3 4 Multi-parent crossover: http://citeseerx.ist.psu.edu/

More information

Implementation of Genetic Algorithm for Agriculture System

Implementation of Genetic Algorithm for Agriculture System Implementation of Genetic Algorithm for Agriculture System Shweta Srivastava Department of Computer science Engineering Babu Banarasi Das University,Lucknow, Uttar Pradesh, India Diwakar Yagyasen Department

More information

Genetic Algorithm: An Optimization Technique Concept

Genetic Algorithm: An Optimization Technique Concept Genetic Algorithm: An Optimization Technique Concept 1 Uma Anand, 2 Chain Singh 1 Student M.Tech (3 rd sem) Department of Computer Science Engineering Dronacharya College of Engineering, Gurgaon-123506,

More information

Genetic Algorithm with Upgrading Operator

Genetic Algorithm with Upgrading Operator Genetic Algorithm with Upgrading Operator NIDAPAN SUREERATTANAN Computer Science and Information Management, School of Advanced Technologies, Asian Institute of Technology, P.O. Box 4, Klong Luang, Pathumthani

More information

College of information technology Department of software

College of information technology Department of software University of Babylon Undergraduate: third class College of information technology Department of software Subj.: Application of AI lecture notes/2011-2012 ***************************************************************************

More information

Genetic Algorithms using Populations based on Multisets

Genetic Algorithms using Populations based on Multisets Genetic Algorithms using Populations based on Multisets António Manso 1, Luís Correia 1 1 LabMAg - Laboratório de Modelação de Agentes Faculdade de Ciências da Universidade de Lisboa Edifício C6, Piso

More information

Conceptual Frameworks for Artificial Immune Systems

Conceptual Frameworks for Artificial Immune Systems Int. Journ. of Unconventional Computing, Vol. 1, pp. 00 00 Reprints available directly from the publisher Photocopying permitted by license only 2005 Old City Publishing, Inc. Published by license under

More information

Agent-Based Architecture of Selection Principle in the Immune System

Agent-Based Architecture of Selection Principle in the Immune System Agent-Based Architecture of Selection Principle in the Immune System Yoshiteru Ishida Graduate School of Information Science Division of Applied Systems Science Nara Institute of Science & Technology Ikoma,

More information

Novel Encoding Scheme in Genetic Algorithms for Better Fitness

Novel Encoding Scheme in Genetic Algorithms for Better Fitness International Journal of Engineering and Advanced Technology (IJEAT) Novel Encoding Scheme in Genetic Algorithms for Better Fitness Rakesh Kumar, Jyotishree Abstract Genetic algorithms are optimisation

More information

In order to have GA, you must have a way to rate a given solution (fitness function). The fitness function must be continuous.

In order to have GA, you must have a way to rate a given solution (fitness function). The fitness function must be continuous. Disclaimer This document is a summary of Prof. Floreano s Bio-inspired Adaptive Machines course. The purpose is to help the student revise for the oral examination. This document should not be considered

More information

Introduction Evolutionary Algorithm Implementation

Introduction Evolutionary Algorithm Implementation Introduction Traditional optimization methods fail when there are complex, nonlinear relationships between the parameters and the value to be optimized, the goal function has many local extrema, and resources

More information

Artificial Homeostasis: Integrating Biologically Inspired Computing

Artificial Homeostasis: Integrating Biologically Inspired Computing Artificial Homeostasis: Integrating Biologically Inspired Computing Jon Timmis Computing Laboratory, University of Kent, Canterbury. UK. Mark Neal Department of Computer Science, University of Wales, Aberystwyth.

More information

Evolutionary Computation

Evolutionary Computation Evolutionary Computation Evolution and Intelligent Besides learning ability, intelligence can also be defined as the capability of a system to adapt its behaviour to ever changing environment. Evolutionary

More information

10. Lecture Stochastic Optimization

10. Lecture Stochastic Optimization Soft Control (AT 3, RMA) 10. Lecture Stochastic Optimization Genetic Algorithms 10. Structure of the lecture 1. Soft control: the definition and limitations, basics of epert" systems 2. Knowledge representation

More information

The Biological Basis of the Immune System as a Model for Intelligent Agents

The Biological Basis of the Immune System as a Model for Intelligent Agents The Biological Basis of the Immune System as a Model for Intelligent Agents Roger L. King 1, Aric B. Lambert 1, Samuel H. Russ 1, and Donna S. Reese 1 1 MSU/NSF Engineering Research Center for Computational

More information

Negative Selection Algorithm :A Survey

Negative Selection Algorithm :A Survey Negative Selection Algorithm :A Survey Delona C Johny,Haripriya P V Department of Information Technology Govt.Engineering college Bartonhill, Trivandrum Anju J S Assistant Professor Department of Information

More information

Mrs. Shahana Gajala Qureshi 1, Mrs. Uzma Arshi Ansari 2

Mrs. Shahana Gajala Qureshi 1, Mrs. Uzma Arshi Ansari 2 IOSR Journal of Computer Engineering (IOSRJCE) ISSN: 2278-0661 Volume 4, Issue 1 (Sep-Oct. 2012), PP 06-13 An efficient and powerful advanced algorithm for solving real coded numerical optimization problem:

More information

Genetic Algorithms in Matrix Representation and Its Application in Synthetic Data

Genetic Algorithms in Matrix Representation and Its Application in Synthetic Data Genetic Algorithms in Matrix Representation and Its Application in Synthetic Data Yingrui Chen *, Mark Elliot ** and Joe Sakshaug *** * ** University of Manchester, yingrui.chen@manchester.ac.uk University

More information

Comp215: Genetic Algorithms - Part 1

Comp215: Genetic Algorithms - Part 1 Comp215: Genetic Algorithms - Part 1 Mack Joyner, Dan S. Wallach (Rice University) Copyright 2016, Mack Joyner, Dan S. Wallach. All rights reserved. Darwin s Theory of Evolution Individual organisms differ

More information