Presentation

479 views

Published on

0 Comments
1 Like
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
479
On SlideShare
0
From Embeds
0
Number of Embeds
38
Actions
Shares
0
Downloads
0
Comments
0
Likes
1
Embeds 0
No embeds

No notes for slide

Presentation

  1. 1. Education and Competences<br />AssociateProfessor<br />Ph.D.Computer Science<br />Systems Engineer<br />Microsoft Certified<br />Developer<br />Paris Sud<br />UniversityOrsay<br />(France)<br />MoscowMicrosoftTraining Centre“Specialist”(Russia)<br />Ecole des Mines de Paris<br />(France)<br />Moscow State Technical Universityof Bauman<br />(Russia)<br /><ul><li>First-class Honours
  2. 2. Silver Decoration
  3. 3. Honour Diploma
  4. 4. First-class Honours
  5. 5. Summa Cum Laude
  6. 6. Qualification Section 27 Computer Science
  7. 7. MCP + MCAD pro excellence certification</li></li></ul><li>Research Field<br />numerical methodsfor optimization<br />Key Objectives :<br /><ul><li>incorporation of intelligence
  8. 8. convergence increasing
  9. 9. constraints handling
  10. 10. multi-criteria optimization
  11. 11. optimization engineering
  12. 12. visualization
  13. 13. benchmarking</li></ul>population-basedapproach<br />Examples :<br /><ul><li> genetic algorithms
  14. 14. differential evolution
  15. 15. evolution strategies</li></ul>Applications :<br />Challenges :<br /><ul><li>multi-criteria
  16. 16. constraints
  17. 17. mixed-variables
  18. 18. design
  19. 19. control
  20. 20. chemistry
  21. 21. time-variable
  22. 22. noisiness
  23. 23.
  24. 24. data mining
  25. 25. scheduling
  26. 26. bioinformatics …</li></li></ul><li>Scientific Contribution Publications<br />Monograph<br />12 conference proceedings with reading committee<br />7 international<br />5 national<br /> 4 national journals with reading committee<br /> 2 book chapters<br />international + national<br /> 3 technical reports<br />
  27. 27. Scientific Contribution<br />EnergeticApproach<br />HybridSVM<br />Originalityof Works<br />UniqueFormula<br />BestResults<br />Transversal<br />Evolution<br />
  28. 28. Valorization<br />Creator at EMA<br />Consultancy<br /><ul><li>Founder of innovative project at the Incubator of EMA
  29. 29. President of scientific association</li></ul>aimed to promote and favour raising of numerical optimization and modern programming technologies in France as well as at the international level<br />
  30. 30. Responsibility and Management<br />Director of innovative project<br />Scientific activity<br />
  31. 31. 200h<br />Teaching<br />Global Optimization<br />Exploratory Data Analysis<br />Random Number Generators<br />Artificial Neural Networks<br />Fuzzy Logic<br />Classification methods<br />Support Vector Machine<br />10 mini missions<br />1st year<br />AssistantProfessor<br />96h<br />Master Info TD<br />DEUG MIAS TD/TP<br />Advanced algorithms :<br /> Linear Programming<br /> Simplex, IPM<br />Programming in C<br />Associate<br />Professor<br />60h<br />3 x 175h<br />Simulation<br />ApproximationBenchmarking<br />Administration of the Enterprise<br />Market Analysis<br />Innovative products creation<br />3 mini missions<br />2nd year<br />3 long projects<br />2nd, 3d et 4th years<br />ProjectDirector<br />
  32. 32. International Relationsand Networks<br />
  33. 33. The step inside<br />
  34. 34. Differential Evolution<br />Part 1<br />
  35. 35. Numerical Optimization<br />Aiming at the best is one of the most fundamental traits of intelligence<br />In all activities human beings tend to<br />Maximize Benefit<br />Minimize Inconvenience<br />Math Optimizationis a collection of Powerful Tools (methods & algorithms) for tackling these real-world problems<br />
  36. 36. Optimization trends<br />from Exact to<br />from Special to<br />from Local to<br />from Individual to<br />from Tuning to<br />Approximate methods<br />Universal solvers<br />Global solutions<br />Collective intelligence<br />Auto-adaptation action<br />MH<br />All this is included inDifferential Evolution<br />
  37. 37. Metaheuristic Optimization<br />Random Optimization<br />Iterative Local Search<br />Swarm Intelligence<br />Evolutionary Computation<br />Simulated Annealing<br />Tabu Search<br />Particle Swarm Optimization<br />Ant Colony Optimization<br />Genetic Algorithms<br />Evolutionary Strategies<br />DE<br /><ul><li> Social Intelligence
  38. 38. Evolution Principles
  39. 39. Physical Laws</li></ul>Differential Evolutioninherits several metaheuristics<br />
  40. 40. Great break-through in Evolutionary Computation<br />Success of DE resides in the manner of the potential solution creation<br />Intelligent use of differences between current solutions realized in a simple and fast linear operatormakes DE unique<br />Concept of DE is a spontaneousself-adaptability to the function<br />best results<br />
  41. 41. Models that can be solved by DE<br />Nonlinear<br />Combinatorial<br />In mixed variables<br />Highly Constrained<br />Multi-modal<br />Multi-objective<br />DE advantages :<br /><ul><li> global optimum
  42. 42. excellent precision
  43. 43. fast convergence
  44. 44. self-adaptation</li></ul>x<br />F(x)<br />Black Box<br />+ Only 0-order information required !<br />
  45. 45. My Contribution to DE<br />Introduction of the Universal Formula of differentiation<br />Classification of the search strategies(random / directed / local / hybrid)<br />Uncovering of the transversal DE species<br />Universalization of the algorithm<br />Development of the energetic selection approach<br />Hybridization DE with regression methods (SVM)<br />Suggestion of new algorithm performance measures<br />Analysis and generalization of some other methods via DE<br />Application in decision making and engineering design<br />
  46. 46. And what DE became now<br />
  47. 47. How does it work ?<br />The simplest example of Differentiation<br />And its general form<br />
  48. 48. - Philosophy Changing<br />- 3 levels of improvement<br />- Search Strategies<br />- Differentiation Analysis<br />- Transversal DE<br />- Some Analogy<br />- Energetic Selection<br />- SVM Hybrid DE<br />Inside ofDifferential Evolution<br />
  49. 49. DE discovers the best solutions<br />Engineering design<br />Scheduling<br />Control<br />Decision-making<br />Image processing<br />Neural networksand Fuzzy systems<br />Chemical engineering and Biosystems<br />Bioinformatics, Computational chemistry and Molecular biology<br />
  50. 50. I solved 2 challenges with DE<br />2. Engineering design<br />“Bump” – a very hard aeronautical benchmark<br />1. Decision-making<br />Identification withthe Choquet Integral<br />Best-known Results !!!<br />
  51. 51. Created Optimization software<br />Part 2<br />
  52. 52. Do best wines with OPTIVINA<br />The answer to wineries’ needs<br /><ul><li>Best possible usage of the vine grapes
  53. 53. Respect of the norms, production and business constraints
  54. 54. Rapid and efficient planning of the vintages and production
  55. 55. Real-time simulation of wine blending
  56. 56. Consider more parameters in wine design
  57. 57. Diversify wine-makers remuneration
  58. 58. Forecast segmentation of production
  59. 59. Better fit production to market needs</li></li></ul><li>VitaEVOLUTION SDK<br />Some screenshots<br />
  60. 60. VitaEVOLUTION SDK<br />Advantages : <br />Software independent<br />3D visualization included<br />Flexible, multi language<br />Extensible for new algorithms<br />Modular for extra packages<br />Use modern technology<br />Has real-world examples<br />Reliable in use<br />Full traceability of actions<br />Surety of results<br />High-quality code<br />Flexible reporting services<br />Compatibility with industrial std<br />Creativity, multi GUI<br />Rapid getting-started<br />Easy and fast programming<br />Warranty service assured<br />and many others …<br />
  61. 61. Web Platform VitaSCIENCES<br />From communication tools toworld lead reference in metaheuristics<br />Standard Library<br />Algorithms & Problems<br />ComputingInterface<br />Problems<br />Algorithms<br />Communication<br />Tools<br />
  62. 62. Web Platform VitaSCIENCES<br />Data Bank<br />Solvers & Problems<br /><ul><li>Compare algorithms
  63. 63. Solve some problem
  64. 64. Add your own elements
  65. 65. Ask for suggestions
  66. 66. Describe a scientific work
  67. 67. Communicate</li></ul>Communication Tools<br />Forum + Chat + VS Space<br />What can we do with VS ?<br />Reporting Services<br /><ul><li>See best results
  68. 68. Order a report
  69. 69. Read news</li></ul>For whom ?<br />researchers PhD studentsprofessorsengineerspublic and private laboratories<br />What Fields :<br />MathematicsComputer ScienceBioinformaticsChemistryLogistics …<br />
  70. 70. Web Platform VitaSCIENCES<br />Everybody can findseveral advantages !<br />Researchers and students :<br /><ul><li>Win a challenge
  71. 71. Augment your experience
  72. 72. Enjoy the space of collaboration and publications
  73. 73. It is your source of inspiration
  74. 74. Profitcollections of algorithms and models
  75. 75. Valorize your competences to find the best job</li></ul>Universities :<br /><ul><li>Demonstrate your performance
  76. 76. Approach the industrials
  77. 77. Application of scientific knowledge to real-world problems
  78. 78. Practice works for students
  79. 79. Space of information exchange
  80. 80. Evaluate your algorithms on-line</li></ul>Industrials :<br /><ul><li>Resolve your problems
  81. 81. Choose the best specialist
  82. 82. Best solutions at reduced price</li></li></ul><li>Thank you for your attention !<br />

×