First National Conference on
           Algorithms and
  Intelligent Systems under Lead
College Activity Shivaji University.
                In association with

Computer Society on India and International Journal of
          Computer and Communication
                   Technology.
Paper Title
 Immune Artificial Cuckoo Search algorithm for
      Nonlinear System Identification.

                Paper ID – 45

     Ritesh kumar                    Lav Kedia
School of electrical sciences   Electrical Engineering
  IIT Bhubaneswar                  BIT Deoghar
CONTENTS…
 Keywords
 FLANN

 Problem Statement

 Cuckoo Search Algorithm

 Introduction to Immunity and its Need

 Advantages

 Pseudocode

 Conclusion
.




    Immune Artificial Cuckoo       Search
    algorithm      for Nonlinear   System
    Identification
KEYWORDS
 FLANN
 NON-LINEAR SYSTEM

 CUCKOO SEARCH ALGORITHM

 IMMUNE SYSTEM
FLANN (FUNCTIONALLY LINKED
ARTIFICIAL NEURAL NETWORK)
 FLANN was first proposed by Pao as a single layer
  ANN structure capable of forming arbitrarily wide
  complex decision regions by generating non-linear
  decision boundaries.
 FLANN has only input and output layers and the
  hidden layers are completely replaced by the
  nonlinear mappings.
 FLANN increases the learning rate but also
  reduces the computational complexity.
CONTD…
Non-Linear plant   Trigonometric FLANN
PROBLEM
To frame algorithm that would be able to model the
  non-linear systems what we encounter in our day-2-
  day life.
NEST

CUCKOO BREEDING




                  eggs
                  Cuckoo
OR




     CROW
LEVY FLIGHTS
   Random walk in which the step-lengths are
    distributed according to a heavy-tailed probability
    distribution. After a large number of steps, the
    distance from the origin of the random walk tends to
    a stable distribution.
IMMUNE ACSO: A BETTER
APPROACH TOWARDS
OPTIMISATION
 Introducing the concept of immunity in AFSO is a
  novel way of development of new algorithms using
  natural swarm intelligence along with the self-
  sustaining capability of immune systems.
 There are several ways of making any optimisation
  algorithm immune. These include Negative
  selection algorithm, clonal selection algorithm and
  Immune Network algorithm.
 For our purpose, we have used clonal selection
  algorithm.
CLONAL SELECTION
 Widely accepted model for how the IMMUNE
  SYSTEM responds to an invasion from outside.
 Some of its basic steps are as follows:

   1. For an attack from outside there is always a
  particular type of receptor cell for detecting it.
   2. After an attack specific type of receptor cells are
  activated and are differentiated to form more
  number of identically specific cells.
   3. So for an attack of similar type these receptor
  cells will come into function quite early and hence
  reduces the risk of cells damage.
CLONAL SELECTION
IN PICTURE
ADVANTAGES OF IMMUNE
AFCSO
 Immune AFSO has a better convergence rate as
  compared to its non-immune counterpart.
 The probability of getting stuck in a local solution
  also gets minimised.
PSEUDOCODE
Begin
Objective function f(x)=(X1,X2,...........)
Generate initial population of n host nests Xi
While (t<max.generation) or (stop criterion)
Get a cuckoo randomly by Levy Flights.
Evaluate its Fitness,Fi.
Choose a nest randomly say j .
If (Fi>Fj)
Replace J by the new solution
End if
A fraction (pA) of the worst nests are abandoned and new ones are
built.
Keep the best solution.
Rank the solutions and find the current best
End while
Use genetic algorithm based crossover and mutation technique to
find the solution to the problems.
Post process results and visualisation.
End.
REFERENCES
   X.-S. Yang; S. Deb (December 2009). "Cuckoo search via Lévy
    flights". World Congress on Nature & Biologically Inspired Computing
    (NaBIC 2009). IEEE Publications. pp. 210–214.
   R. N. Mantegna, Fast, accurate algorithm for numerical simulation of
    Levy stable stochastic processes, Physical Review E, Vol.49, 4677-
    4683 (1994).
   J.J.,1989a.AdaptiveIIRfiltering.IEEEASSPMagazine,421.
    1989,chapter 8,pp.197-222.
   X.-S. Yang and S. Deb, "Engineering optimisation by cuckoo search",
    Int. J. Mathematical Modelling and Numerical Optimisation", Vol. 1,
    No. 4, 330-343 (2010).
   N. Bacanin, An object-oriented software implementation of a novel
    cuckoo search algorithm, Proc. of the 5th European Conference on
    European Computing Conference (ECC'11), pp. 245-250 (2011).
CONTD…
   M. Tuba, M. Subotic, and N. Stanarevic, Modified cuckoo search
    algorithm for unconstrained optimization problems, Proc. of the 5th
    European Conference on European Computing Conference
    (ECC'11), pp. 263-268 (2011).
   S. Walton, O. Hassan, K. Morgan, Using proper orthogonal
    decomposition to reduce the order ot optimization problems, in: Proc.
    16th Int. Conf. on Finite Elments in Flow Problems (Eds. Wall W.A.
    and Gvravemeier V.), Munich, p.90 (2011).
   F. Wang, L. Lou, X. He, Y. Wang, Hybrid optimization algorithm of
    PSO and Cuckoo Search, in: Proc. of 2nd Int. Conference on
    Artificial Intelligence, Management Science and Electronic
    Commerce (AIMSEC'11), pp. 1172-1175 (2011).
   A. Kumar and S. Chakarverty, Design optimization for reliable
    embedded system using Cuckoo Search,in: Proc. of 3rd Int.
    Conference on Electronics Computer Technology (ICECT2011), pp.
    564-268 (2011).
BRAIN TEASERS…
 So which famous algorithm is modified in Cuckoo
  Search algorithm??
 So can anyone tell me that who was the first person to
  work on evolutionary computation??
 Why do we prefer evolutionary algorithms over others to
  solve or optimize our problems??
 So what all shortcomings of PSO are overcome by
  Cuckoo search algo.??
Cuckoo search algorithm

Cuckoo search algorithm

  • 1.
    First National Conferenceon Algorithms and Intelligent Systems under Lead College Activity Shivaji University. In association with Computer Society on India and International Journal of Computer and Communication Technology.
  • 2.
    Paper Title ImmuneArtificial Cuckoo Search algorithm for Nonlinear System Identification. Paper ID – 45 Ritesh kumar Lav Kedia School of electrical sciences Electrical Engineering IIT Bhubaneswar BIT Deoghar
  • 3.
    CONTENTS…  Keywords  FLANN Problem Statement  Cuckoo Search Algorithm  Introduction to Immunity and its Need  Advantages  Pseudocode  Conclusion
  • 4.
    . Immune Artificial Cuckoo Search algorithm for Nonlinear System Identification
  • 5.
    KEYWORDS  FLANN  NON-LINEARSYSTEM  CUCKOO SEARCH ALGORITHM  IMMUNE SYSTEM
  • 6.
    FLANN (FUNCTIONALLY LINKED ARTIFICIALNEURAL NETWORK)  FLANN was first proposed by Pao as a single layer ANN structure capable of forming arbitrarily wide complex decision regions by generating non-linear decision boundaries.  FLANN has only input and output layers and the hidden layers are completely replaced by the nonlinear mappings.  FLANN increases the learning rate but also reduces the computational complexity.
  • 7.
  • 8.
    Non-Linear plant Trigonometric FLANN
  • 9.
    PROBLEM To frame algorithmthat would be able to model the non-linear systems what we encounter in our day-2- day life.
  • 11.
  • 12.
    OR CROW
  • 13.
    LEVY FLIGHTS  Random walk in which the step-lengths are distributed according to a heavy-tailed probability distribution. After a large number of steps, the distance from the origin of the random walk tends to a stable distribution.
  • 14.
    IMMUNE ACSO: ABETTER APPROACH TOWARDS OPTIMISATION  Introducing the concept of immunity in AFSO is a novel way of development of new algorithms using natural swarm intelligence along with the self- sustaining capability of immune systems.  There are several ways of making any optimisation algorithm immune. These include Negative selection algorithm, clonal selection algorithm and Immune Network algorithm.  For our purpose, we have used clonal selection algorithm.
  • 15.
    CLONAL SELECTION  Widelyaccepted model for how the IMMUNE SYSTEM responds to an invasion from outside.  Some of its basic steps are as follows: 1. For an attack from outside there is always a particular type of receptor cell for detecting it. 2. After an attack specific type of receptor cells are activated and are differentiated to form more number of identically specific cells. 3. So for an attack of similar type these receptor cells will come into function quite early and hence reduces the risk of cells damage.
  • 16.
  • 17.
    ADVANTAGES OF IMMUNE AFCSO Immune AFSO has a better convergence rate as compared to its non-immune counterpart.  The probability of getting stuck in a local solution also gets minimised.
  • 18.
    PSEUDOCODE Begin Objective function f(x)=(X1,X2,...........) Generateinitial population of n host nests Xi While (t<max.generation) or (stop criterion) Get a cuckoo randomly by Levy Flights. Evaluate its Fitness,Fi. Choose a nest randomly say j . If (Fi>Fj) Replace J by the new solution End if A fraction (pA) of the worst nests are abandoned and new ones are built. Keep the best solution. Rank the solutions and find the current best End while Use genetic algorithm based crossover and mutation technique to find the solution to the problems. Post process results and visualisation. End.
  • 19.
    REFERENCES  X.-S. Yang; S. Deb (December 2009). "Cuckoo search via Lévy flights". World Congress on Nature & Biologically Inspired Computing (NaBIC 2009). IEEE Publications. pp. 210–214.  R. N. Mantegna, Fast, accurate algorithm for numerical simulation of Levy stable stochastic processes, Physical Review E, Vol.49, 4677- 4683 (1994).  J.J.,1989a.AdaptiveIIRfiltering.IEEEASSPMagazine,421. 1989,chapter 8,pp.197-222.  X.-S. Yang and S. Deb, "Engineering optimisation by cuckoo search", Int. J. Mathematical Modelling and Numerical Optimisation", Vol. 1, No. 4, 330-343 (2010).  N. Bacanin, An object-oriented software implementation of a novel cuckoo search algorithm, Proc. of the 5th European Conference on European Computing Conference (ECC'11), pp. 245-250 (2011).
  • 20.
    CONTD…  M. Tuba, M. Subotic, and N. Stanarevic, Modified cuckoo search algorithm for unconstrained optimization problems, Proc. of the 5th European Conference on European Computing Conference (ECC'11), pp. 263-268 (2011).  S. Walton, O. Hassan, K. Morgan, Using proper orthogonal decomposition to reduce the order ot optimization problems, in: Proc. 16th Int. Conf. on Finite Elments in Flow Problems (Eds. Wall W.A. and Gvravemeier V.), Munich, p.90 (2011).  F. Wang, L. Lou, X. He, Y. Wang, Hybrid optimization algorithm of PSO and Cuckoo Search, in: Proc. of 2nd Int. Conference on Artificial Intelligence, Management Science and Electronic Commerce (AIMSEC'11), pp. 1172-1175 (2011).  A. Kumar and S. Chakarverty, Design optimization for reliable embedded system using Cuckoo Search,in: Proc. of 3rd Int. Conference on Electronics Computer Technology (ICECT2011), pp. 564-268 (2011).
  • 21.
    BRAIN TEASERS…  Sowhich famous algorithm is modified in Cuckoo Search algorithm??  So can anyone tell me that who was the first person to work on evolutionary computation??  Why do we prefer evolutionary algorithms over others to solve or optimize our problems??  So what all shortcomings of PSO are overcome by Cuckoo search algo.??