Evolutionary Algorithms for Self-Organising Systems
Upcoming SlideShare
Loading in...5
×
 

Like this? Share it with your network

Share

Evolutionary Algorithms for Self-Organising Systems

on

  • 1,265 views

Talk I gave at Ben Gurion University of the Negev in Israel on the 24rd/June/2009. These are a series of talks for the period in which I visited BGU as a distinguished visiting scientist

Talk I gave at Ben Gurion University of the Negev in Israel on the 24rd/June/2009. These are a series of talks for the period in which I visited BGU as a distinguished visiting scientist

Statistics

Views

Total Views
1,265
Views on SlideShare
1,260
Embed Views
5

Actions

Likes
0
Downloads
35
Comments
0

1 Embed 5

http://www.cs.nott.ac.uk 5

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

Evolutionary Algorithms for Self-Organising Systems Presentation Transcript

  • 1. An Evolutionary Algorithm Approach to Guiding the Evolution of Self-Organised Systems Natalio Krasnogor Interdisciplinary Optimisation Laboratory Automated Scheduling, Optimisation & Planning Research Group School of Computer Science Centre for Integrative Systems Biology School of Biology Centre for Healthcare Associated Infections Institute of Infection, Immunity & Inflammation University of Nottingham Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 1 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 2. Previous Talk Slides At http://www.slideshare.net/nxk Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 2 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 3. Overview • Motivation • Towards “Dial a Pattern” in Complex Systems • Methodological Overview • Virtual Complex Systems Au • Physical Complex Systems • Nanoparticle Simulation Details • Results • Conclusions & Further work Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 3 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 4.  This work was done in collaboration with Prof. P. Moriarty and his group at the School of Physics and Astronomy at the University of Nottingham  Based on the papers P.Siepmann, C.P. Martin, I. Vancea, P.J. Moriarty, and N. Krasnogor. A genetic algorithm approach to probing the evolution of self-organised nanostructured systems. Nano Letters, 7(7):1985-1990, 2007. http://dx.doi.org/10.1021/nl070773m G. Terrazas, P. Siepman, G. Kendal, and N. Krasnogor. An evolutionary methodology for the automated design of cellular automaton-based complex systems. Journal of Cellular Automata, 2(1):77-102, 2007. http://www.oldcitypublishing.com/JCA/JCA.html L. Cronin, N. Krasnogor, B. G. Davis, C. Alexander, N. Robertson, J.H.G. Steinke, S.L.M. Schroeder, A.N. Khlobystov, G. Cooper, P. Gardner, P. Siepmann, and B. Whitaker. The imitation game—a computational chemical approach to recognizing life. Nature Biotechnology, 24:1203-1206, 2006. All papers available at: http://www.cs.nott.ac.uk/~nxk/publications.html Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 4 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 5. Motivation - Automated design and optimisation of complex systems’ target behaviour - cellular automata/ ODEs/ P-systems models - physically/chemically/biologically implemented -present a methodology to tackle this problem -supported by experimental illustration Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 5 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 6. Major advances in the rational/analytical design of large and complex systems have been reported in the literature and more recently the automated design and optimisation of these systems by modern AI and Optimisation tools have been introduced. It is unrealistic to expect every large & complex physical, chemical or biological system to be amenable to hand-made fully analytical designs/optimisations. We anticipate that as the number of research challenges and applications in these domains (and their complexity) increase we will need to rely even more on automated design and optimisation based on sophisticated AI & machine learning Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 6 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 7. Major advances in the rational/analytical design of large and complex systems have been reported in the literature and more recently the automated design and optimisation of these systems by This has happened before in other modern AI and Optimisation tools have been introduced. research and industrial disciplines,e.g: It is unrealistic to expect every large & complex physical, chemical •VLSI design or biological system to be amenable to hand-made fully analytical •Space antennae design designs/optimisations. •Transport Network design/optimisation •Personnel Rostering •Scheduling and timetabling We anticipate that as the number of research challenges and applications in these domains (and their complexity) increase we will need to rely even more on automated design and optimisation based on sophisticated AI & machine learning Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 6 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 8. Major advances in the rational/analytical design of large and complex systems have been reported in the literature and more with That is, complex systems are plagued NP-Hardness, non-approximability, recently the automated design and optimisation of these systems by modern AI and Optimisation toolsuncertainty, undecidability, etc results This has happened before in other have been introduced. research and industrial disciplines,e.g: It is unrealistic to expect every large & complex physical, chemical •VLSI design or biological system to be amenable to hand-made fully analytical •Space antennae design designs/optimisations. •Transport Network design/optimisation •Personnel Rostering •Scheduling and timetabling We anticipate that as the number of research challenges and applications in these domains (and their complexity) increase we will need to rely even more on automated design and optimisation based on sophisticated AI & machine learning Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 6 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 9. Major advances in the rational/analytical design of large and complex systems have been reported in the literature and more with That is, complex systems are plagued NP-Hardness, non-approximability, recently the automated design and optimisation of these systems by modern AI and Optimisation toolsuncertainty, undecidability, etc results This has happened before in other have been introduced. research and industrial disciplines,e.g: It is unrealistic to expect every large & complex physical, chemical •VLSI design or biological system to be amenable to hand-made fully analytical •Space antennae design designs/optimisations. •Transport Network design/optimisation •Personnel Rostering Yet, they are routinely solved by •Scheduling and timetabling We anticipate that as the number of research challenges and design sophisticated optimisation and techniques, like evolutionary applications in these domains (and their complexity) increase we algorithms, machine learning, etc will need to rely even more on automated design and optimisation based on sophisticated AI & machine learning Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 6 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 10. Automated Design/Optimisation is not only good because it can solve larger problems but also because this approach gives access to different regions of the space of possible designs (examples of this abound in the literature) Space of all possible designs/optimisations Automated Analytical Design Design (e.g. evolutionary) A distinct view of the space of possible designs could enhance the understanding of underlying system Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 7 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 11. The research challenge :  For the Engineer, Chemist, Physicist, Biologist :  To come up with a relevant (MODEL) SYSTEM M*  For the Computer Scientist:  To develop adequate sophisticated algorithms -beyond exhaustive search- to automatically design or optimise existing designs on M* regardless of computationally (worst-case) unfavourable results of exact algorithms. Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 8 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 12. Towards “Dial a Pattern” in Complex Systems Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 9 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 13. Towards “Dial a Pattern” in Complex Systems es ctur Stru ical .S Lex C te re rete isc D ted Disc u st rib Di Continuous (simulated) CS How do we program? Disc rete /Con tin. ( phys ical) CS Dis cre te/C ont inu os (B iolo gic al) Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 9 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 14. Methodological Overview Dial a Pattern requires:  Parameter Learning/Evolution Technology  Structural Learning/Evolution Technology  Integrated Parameter/Structural Learning/Evolution Tech.  in silico or experimental implementation Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 10 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 15. Initial Attempts at a “Dial a Pattern” Methodology behaviour CA-based / Real emergent vs target complex system Parameters/Structure Evolutionary algorithms Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 11 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 16. Embodied Evolution  Evolutionary Scheme  Some parts of it are embedded into a physical, chemical or biological substrate. Strong embodiment Week embodiment Genes Phenotypes Fitnesses Variation & selection mechanisms (or other metaheuristic scheme) Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 12 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 17. A Complex Mapping Fitness(es) Phenotypes Genotypes Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 13 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 18. The CHELLnet: Unifying Investigation in Artificial Cellularity and Complexity Synthesis of abiotic life-like functionality in complex chemical systems through open-ended evolution The CHELLnet comprised four sub-projects, each with researchers in universities across the UK http://www.CHELLnet.org Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 14 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 19.  Life-like functionality through evolved complexity in 3 different platforms Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 15 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 20. What is the CHELLnet? BrainCHELL - directing assembly of conducting networks so that there is function encoded in the structure of the product. Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 16 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 21. What is the CHELLnet? VesiCHELL - complexity and pattern formation within lipid-bounded systems Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 17 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 22. What is the CHELLnet? WellCHELL - model miniature laboratory system with multiple chemical flow reactors where conditions of chemical processes computer controlled Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 18 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 23. What is the CHELLnet? Evolvable CHELLware wellCHELL behaviour brainCHELL emergent vs target vesiCHELL CHELL platforms Evolutionary algorithms parameters Evolvable CHELLware Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 19 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 24. “we will implement an object-oriented, platform-independent, evolutionary engine (EE). The EE will have a user-friendly interface that will allow the various platform users (i.e. wellCHELL, The Evolutionary Engine brainCHELL, vesiCHELL) to specify the platform with which the EE will interact” Evolvable CHELLware grant application - no data types - no evaluation module - data types and bounds  - no parameters - evaluation module (‘plug in’)  - EA or other ML parameters  specialised generic GA results GA XML Evaluation module Java servlet problem-specific web-based web-based configuration execution module module Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 20 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 25. What is the CHELLnet? Evolvable CHELLware Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 21 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 26. What is the CHELLnet? Evolvable CHELLware Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 22 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 27. What is the CHELLnet? Evolvable CHELLware Log details Results graph Visual Visual representation representation of target of best result if applicable Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 23 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 28. What is the CHELLnet? Evolvable CHELLware Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 24 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 29. What is the CHELLnet? Evolvable CHELLware First steps towards embodied evolution on multiple, distinct platforms. This are being developed. We have proofs of concept working with models/simulators: 1.Proof of concept using cellular automaton-based models 2.Self-organised nanostructured systems 3. WellChell (in Manchester) 4. SPM (in Nottingham, 2 sites [CS, P&A]) Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 25 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 30. Examples of Target Evolution in Complex systems Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 26 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 31. Parameter Learning/Evolution Technology Example - Self-organising processes - Modelled using cellular automata, gass latice, ODEs, etc - infinite, regular grid of cells - each cell in one of a finite number of states - at a given time, t, the state of a cell is a function of the states of its neighbourhood at time t-1. Example - infinite sheet of graph paper - each square is either black or white ? - in this case, neighbours of a cell are the eight squares touching it - for each of the 28 possible patterns, a rules table would state whether the center cell will be black or white on the next time step. Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 27 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 32. CA continuous Turbulence Gas Lattice Ben-Gurion University of the Negev Distinguished Scientist Visitor Program Gas Lattice 28 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 33. CA continuous Turbulence Gas Lattice d ve n ol ive Ev G globals [ row ;; current row we are now calculating done? ;; flag used to allow you to press the go button multiple times ] patches-own [ value ;; some real number between 0 and 1 ] to setup-general set row screen-edge-y ;; Set the current row to be the top set done? false cp ct end ;; ] end …….. Ben-Gurion University of the Negev Distinguished Scientist Visitor Program Gas Lattice 28 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 34. Structural Learning/Evolution Technology Example Wang Tiles Models Temperature T Glue Strength Matrix Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 29 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 35. Structural Learning/Evolution Technology Example Wang Tiles Models en iv G Temperature T Glue Strength Matrix d ve ol Ev Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 29 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 36. Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 30 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 37. Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 31 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 38. Parameter Learning/Evolution Technology Example lecA- PAO1 mvaT- Env. Params Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 32 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 39. Parameter Learning/Evolution Technology Example lecA- PAO1 mvaT- d d ve ve ol ol Ev Ev Env. Params Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 32 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 40. How Do We Program These Complex Systems? behaviour Complex System emergent vs target How do we measure this? parameters How similar is to ? Evolutionary algorithms Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 33 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 41. The Universal Similarity Metric (USM) is a measure of similarity between two given objects in terms of information distance: where K(o) is the Kolmogorov complexity Prior Kolmogorov complexity K(o): The length of the shortest program for computing o by a Turing machine Conditional Kolmogorov complexity K(o1|o2): How much (more) information is needed to produce object o1 if one already knows object o2 (as input) Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 34 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 42. The Universal Similarity Metric (USM) - Is the USM a good objective function for evolving target spacio-temporal behaviour in a CA system? - methodology for answering this question - experimental results Fitness Distance Correlation GENOTYPE PHENOTYPE FITNESS CA model USM Clustering Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 35 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 43. Data set For each CA system: • Keep all but one parameter the same • Produce 10 behaviour patterns through the variable parameter • Repeat for other parameters EXAMPLE turb_c4 refers to the spacio-temporal pattern produced by the fourth variation in parameter c of a Turbulence CA system Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 36 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 44. Produced by MODEL(p1,p2,…,pn) p1 p2 pn Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 37 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 45. Clustering • does the USM detect similarity of phenotype with a target pattern? • if yes, it should be able to correctly cluster spatio-temporal patterns that look similar together • and, those similar patterns should be related to a specific family of images arising from the variation of a single parameter Fitness Distance Correlation GENOTYPE PHENOTYPE FITNESS CA model USM • calculate a similarity matrix filled with the results Clustering of the application of the USM to a set of objects • during the clustering process, similar objects should be grouped together Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 38 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 46. Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 39 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 47. Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 40 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 48. Fitness Distance Correlation • correlation analyses of a given fitness function versus parametric (genotype) distance. • larger numbers indicate the problem could be optimised by a GA • numbers around zero [-0.15, 0.15] indicate bad correlation • scatter plots are helpful Fitness Distance Correlation GENOTYPE PHENOTYPE FITNESS CA model USM Target Clustering 1 2 3 distance = 2 Fitness = USM (T,D) Designoid 1 4 3 Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 41 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 49. Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 42 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 50. The Evolutionary Engine “we will implement an object-oriented, platform-independent, evolutionary engine (EE). The EE will have a user-friendly interface that will allow the various platform users to specify the platform with which the EE will interact” Evolvable CHELLware grant application - no data types - no evaluation module - data types and bounds  - no parameters - evaluation module (‘plug in’)  - GA parameters  specialised generic GA results GA XML Evaluation module Java servlet problem-specific web-based web-based configuration execution module module Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 43 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 51. A brief overview of Genetic Algorithms Motivation - optimisation problems global optimum - large search space - inspired by Darwinian evolution - area covered? - degree of order? - similarity to target pattern? 22 0.25 1.0 4.5 1.05 simulator fitness function genotype fitness phenotype Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 44 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 52. Results on CAs Target Designoid e5 f3 . Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 45 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 53. Target Designoid Target usm(F,T) e(i) e(c) e(r) E p 0.91980 0.26843 0.35314 0.05552 0.22569 . Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 46 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 54. Dialling a Pattern in Meta-Automata  Remember the standard numbering of rules: Encoding of the elementary rule 145 t0 Neighbourhoods at t3 Output states at t4 Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 47 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 55.  A Meta-Automaton is a special class of non- uniform automata  Its defined by a spatio-temporal lattice  The set of 256 standard rules  Special variables k-cells and t-times  The semantics is:  k consecutive cells are assigned to the same rules, rules can be different among distinct k-groups  Every Total_Time/ t timesteps rules are reassigned to groups Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 48 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 56. Meta-Automaton (k=2, t=2) k=2 Group 1 Group 2 Phase 1 t=2 Phase 2 Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 49 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 57. Evolving (k=1,2,t=1) Meta-Automaton Target Designoid Target Designoid Target Designoid T D T D Target Designoid Target Designoid Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 50 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 58. Evolving (k=4,t=1) Meta-Automaton G. Terrazas, P. Siepman, G. Kendal, and N. Krasnogor. An evolutionary methodology for the automated design of cellular automaton-based complex systems. Journal of Cellular Automata, 2007 Target Designoid Target Designoid Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 51 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 59. Self-Organised Nanostructured Systems Thiol-passivated Au nanoparticles Gold core Thiol groups Au Sulphur ‘head’ Alkane ‘tail’, e.g. octane ~3nm Dispersed in toluene, and spin cast onto native-oxide-terminated silicon Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 52 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 60. Au nanoparticles: Morphology AFM images taken by Matthew O. Blunt, Nottingham Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 53 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 61. Nanoparticle Simulations Solvent is represented as a two- dimensional lattice gas Each lattice site represents 1nm2 Nanoparticles are square, and occupy nine lattice sites Based on the simulations of Rabani et al. (Nature 2003, 426, 271-274). Includes modifications to include next-nearest neighbours to remove anisotropy. http://www.nottingham.ac.uk/physics/research/nano/ Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 54 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 62. Nanoparticle Simulations • The simulation proceeds by the Metropolis algorithm: – Each solvent cell is examined and an attempt is made to convert from liquid to vapour (or vice-versa) with an acceptance probability pacc = min[1, exp(-ΔH/kBT)] – Similarly, the particles perform a random walk on wet areas of the substrate, but cannot move into dry areas. – The Hamiltonian from which ΔH is obtained is as follows: Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 55 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 63. Nanoparticle Simulations Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 56 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 64. Nanoparticle Simulations Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 56 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 65. Nanoparticle Simulations Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 57 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 66. Nanoparticle Simulations Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 57 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 67. Nanoparticle Simulations Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 58 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 68. Nanoparticle Simulations Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 58 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 69. A brief overview of Genetic Algorithms Evolution - Recombination (mating) e.g. exchanging parameters ‘combine the best bits of each parent’ - Mutation e.g. altering the value of a parameter at random with some small probability GENERATION 0 TIME Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 59 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 70. A brief overview of Genetic Algorithms Evolution - Recombination (mating) e.g. exchanging parameters ‘combine the best bits of each parent’ - Mutation e.g. altering the value of a parameter at random with some small probability GENERATION 1 TIME Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 60 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 71. A brief overview of Genetic Algorithms Evolution - Recombination (mating) e.g. exchanging parameters ‘combine the best bits of each parent’ - Mutation e.g. altering the value of a parameter at random with some small probability GENERATION 1 TIME Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 60 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 72. A brief overview of Genetic Algorithms Evolution - Recombination (mating) e.g. exchanging parameters ‘combine the best bits of each parent’ - Mutation e.g. altering the value of a parameter at random with some small probability GENERATION 2 TIME Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 61 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 73. A brief overview of Genetic Algorithms Evolution - Recombination (mating) e.g. exchanging parameters ‘combine the best bits of each parent’ - Mutation e.g. altering the value of a parameter at random with some small probability GENERATION 2 TIME Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 61 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 74. A brief overview of Genetic Algorithms Evolution - Recombination (mating) e.g. exchanging parameters ‘combine the best bits of each parent’ - Mutation e.g. altering the value of a parameter at random with some small probability GENERATION 3 TIME Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 62 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 75. A brief overview of Genetic Algorithms Evolution - Recombination (mating) e.g. exchanging parameters ‘combine the best bits of each parent’ - Mutation e.g. altering the value of a parameter at random with some small probability GENERATION 3 TIME Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 62 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 76. A brief overview of Genetic Algorithms Evolution - Recombination (mating) e.g. exchanging parameters ‘combine the best bits of each parent’ - Mutation e.g. altering the value of a parameter at random with some small probability TIME Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 63 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 77. A brief overview of Genetic Algorithms Evolution - Recombination (mating) e.g. exchanging parameters ‘combine the best bits of each parent’ - Mutation e.g. altering the value of a parameter at random with some small probability converges to optimum solution FITNESS TIME Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 63 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 78. Evolving towards a target pattern (simulated) • Selected a target image from simulated data set • Initialised GA - Roulette Wheel selection - Uniform crossover (probability 1) - Random reset mutation (probability 0.3) - Population size: 10 Target: - Offspring: 5 - µ + λ replacement • Ran the GA for 200 iterations - on a single processor server, run time ≈ 5 days - using Nottingham’s cluster (up to 10 nodes), run time ≈ 12 hours Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 64 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 79. Evolving towards a target pattern (simulated) Evolving to a simulated target Target: 0.960 0.945 Fitness 0.930 Average Best 0.915 0.900 0 2 4 6 8 11 15 19 23 27 31 35 39 43 47 51 55 59 63 67 71 75 79 83 87 91 95 99 104 110 116 122 128 134 140 146 152 158 164 170 176 182 188 194 200 Generations Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 65 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 80. Evolving towards a target pattern (experimental) Evolving to a experimental target Target: 1.000 0.975 Fitness 0.950 Average Best 0.925 0.900 0 3 6 9 13 18 23 28 33 38 43 48 53 58 63 68 73 78 83 88 93 98 104 111 118 125 132 139 146 153 160 167 174 181 188 195 Generations Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 66 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 81.  Using only the same fitness function as for the CAs was not sufficient for matching simulation to experimental data  We extended the image analysis, i.e. fitness function, to Minkowsky functionals, namely, area, perimeter and euler characteristic Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 67 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 82. Self-organising nanostructures Minkowski Functionals Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 68 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 83. Self-organising nanostructures Evolved design: Minkowski functionals Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 69 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 84. Self-organising nanostructures Evolved design: Minkowski functionals Robustness checking Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 70 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 85. Self-organising nanostructures Evolved design: Minkowski functionals Robustness checking: i) Clustering Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 71 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 86. Self-organising nanostructures Evolved design: Minkowski functionals Robustness checking: ii) Fitness Distance Correlation 1/Fitness Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 72 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 87. Self-organising nanostructures Evolved design: Minkowski functionals Robustness checking: ii) Fitness Distance Correlation 1/Fitness Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 73 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 88. Self-organising nanostructures Evolved design: Minkowski functionals Robustness checking: ii) Fitness Distance Correlation 1/Fitness Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 74 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 89. Self-organising nanostructures Experimental target set Cell Island Labyrinth Worm Evolved set Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 75 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 90. Self-organising nanostructures Experimental target set Cell Island Labyrinth Worm Evolved set Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 75 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 91. Self-organising nanostructures Experimental target set Cell Island Labyrinth Worm Evolved set Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 75 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 92. Self-organising nanostructures Experimental target set: Results P.Siepmann, C.P. Martin, I. Vancea, P.J. Moriarty, and N. Krasnogor. A Genetic Algorithm for Evolving Patterns in Nanostructured systems. Nano Letters (to appear) The analysis of the designability of specific patterns is important as some patterns are more evolvable (multiple solutions) than others and Smart surface design Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 76 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 93. Conclusions • We can evolve target simulated behaviour using a GA with the USM but the USM is not enough •For evolving target experimental designs we used Minkowsky functionals (e.g. Area, Perimeter, Euler Characteristics) • Using Fitness Distance Correlation and Clustering, we can show whether a given fitness function is/isn’t an appropriate objective function for a given domain. • Can we generate a target spatio-temporal behaviour in a CA/Real system? YES - GA generates very convincing designoid patterns Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 77 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 94. Future Work (I)  use of more problem-specific fitness functions  open ended (multiobjective) evolution  e.g. “evolve a pattern with as many large spots as possible in as ordered a fashion as possible”  parameter investigations  larger populations  full fitness landscape analysis  Noisy, expensive, multiobjective fitness functions  Datamining the results Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 78 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 95. Future Work (II) Collect Data Evolve models using Evolutionary “reality runs (RR)” results as targets Expensive, noisy, Design for the models themselves Stochastic, etc Evolve parameters to approximate target behaviour of desired system Physical, Chemical, Biological Model System Abstracted into a model, e.g., ODE, NN, “cook book”, etc Evolutionary Design Try best estimates from model parameters Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 79 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 96. Applications (in design and manufacture) and further work - Many, many systems can be modelled using CAs/Monte Carlos -Many complex physical/chemical systems need to be programmed - Research into chemical ‘design’ We are actively working towards these practical goals in the context of the EPSRC grant CHELLnet (EP/D023343/1), which comprises e.g. designoid patterns in the BZ reaction Evolvable CHELLware (EP/D021847/1), vesiCHELL (EP/D022304/1), brainCHELL (EP/D023645/1) and wellCHELL (EP/D023807/1). and self-organising nanostructured systems CHELLNet http://www.chellnet.org Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 80 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009
  • 97. Acknowledgements  Prof. P. Moriarty (School of Physics and Astronomy, UoN)  EPSRC, BBSRC for funding  BGU for funding the DSVP  Specially to Prof. Moshe Sipper for hosting me at BGU!  Any questions? Ben-Gurion University of the Negev Distinguished Scientist Visitor Program 81 /81 Beer Sheva, Israel - 23/5 to 6/7 2009 Thursday, 25 June 2009