• Like
  • Save
Presentation
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

Published

 

  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
No Downloads

Views

Total Views
318
On SlideShare
0
From Embeds
0
Number of Embeds
1

Actions

Shares
Downloads
0
Comments
0
Likes
1

Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide

Transcript

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