Download presentation

Presentation is loading. Please wait.

Published byIsabella Hickey Modified over 3 years ago

1
University of Rostock Institute of Applied Microelectronics and Computer Engineering 1 Mutation at Evolution Strategy by Guido Moritz SoftComputingMethods 2006

2
University of Rostock Institute of Applied Microelectronics and Computer Engineering 2 Target of Evolution Strategy Find a solution for BlackBoxProblems (no explicit solution) wich is exactly enough. INPUTOUTPUT EXAMPLE: FIND AN INPUT WHERE THE OUTPUT IS MAXIMUM

3
University of Rostock Institute of Applied Microelectronics and Computer Engineering 3 Target of Evolution Strategy P I D Proportionalanteil Integralanteil Differentialanteil x (t)x (t) Aktion Reaktion Regelgröße by Ingo Rechenberg

4
University of Rostock Institute of Applied Microelectronics and Computer Engineering 4 Evolution Strategy – how to Genererating new elements by recombination/variation of existing elements Choose good and bad elements (because of difference between OUTPUTS) Take good ones for next generation (recombination/variation) - > creating new INPUTS

5
University of Rostock Institute of Applied Microelectronics and Computer Engineering 5 Evolution Strategy – how to Creating elements randomly Select parents (by random) Recombination of parents Mutation Choose because of fitness Generating new generation Xneu=Xalt+*N(0,σ)

6
University of Rostock Institute of Applied Microelectronics and Computer Engineering 6 Mutation – how to Changing a value by f.e. adding or substracting a small normal distributed (avarage=0) value with a standard variance (dt. standartabweichung) How big changing-decided by and standart variance of N() Xneu=Xalt+*N(0,σ)

7
University of Rostock Institute of Applied Microelectronics and Computer Engineering 7 Mutation – how to GALTONs Nailboard (Nails vertical of wall) by Ingo Rechenberg Leakage=distance between nails

8
University of Rostock Institute of Applied Microelectronics and Computer Engineering 8 Selfadapting Leakage (StepSize) - Why x1 h1 x2 h2 x1=x2 BUT h1!=h2

9
University of Rostock Institute of Applied Microelectronics and Computer Engineering 9 Rechenberg 1/5 Rule If 1/5 of mutations are better (better fitness) decrease leakage! If sucess<1/5 = *1,5; Else if (sucess>1/5) = /1,5; Else = ;

10
University of Rostock Institute of Applied Microelectronics and Computer Engineering 10 Problems Rechenbergs Rule is static and depends not on problem itself (maybe only local optimum) Schwefel enhanced Rechenbergs Rule ( takes part at evolution): σ neu := σ alt e^N(0,Δ) xneu := xalt + *N(0, σ neu) σ can addapt itself to problem Δ-factor how strong is selfadapting of leakage http://www.evocomp.de/themen/evolutionsstrategien/evostrat.html

11
University of Rostock Institute of Applied Microelectronics and Computer Engineering 11 Random Numbers Constant allocated (same chance) Gauß allocated

12
University of Rostock Institute of Applied Microelectronics and Computer Engineering 12 Random Numbers Take quadratic values –Gauß narrow/higher –Constand bigger values Group numbers –Constand getting closer to avarage Effect of both (quadrativ&group) –Difference between values and avarage is getting smaller

Similar presentations

OK

Notes Over 4.8 Identifying Functions A relation where each input has exactly one output. Function Decide whether the relation is a function. If it is.

Notes Over 4.8 Identifying Functions A relation where each input has exactly one output. Function Decide whether the relation is a function. If it is.

© 2017 SlidePlayer.com Inc.

All rights reserved.

Ads by Google

Ppt on managerial economics introduction Ppt on electricity for class 10th book Ppt on statistics and probability formulas Download ppt on android operating system Ppt on bank lending rates Ppt on product specification software Presentation ppt on motivation and emotion Ppt on latest technology free download Ppt on trial and error supernatural Ppt on company act 1956 provisions