1 / 28

Genetic Programming as a Tool for novel Creation CS 621 Seminar

Genetic Programming as a Tool for novel Creation CS 621 Seminar Sri Raj Paul(08305034) Course Instructor Balamurali(08405401) Prof. Pushpak Bhattacharyya. The way we go…. Invention & Patent AI & Invention Genetic Algorithm Genetic Programming GP – Invention Machine

mliss
Download Presentation

Genetic Programming as a Tool for novel Creation CS 621 Seminar

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Genetic Programming as a Tool for novel Creation CS 621 Seminar Sri Raj Paul(08305034) Course Instructor Balamurali(08405401) Prof. Pushpak Bhattacharyya

  2. The way we go…. • Invention & Patent • AI & Invention • Genetic Algorithm • Genetic Programming • GP – Invention Machine • Conclusion

  3. Invention • What is it? a new form, composition of matter, device, or process • What is a patent? a set of exclusive rights granted by a state to an inventor or his assignee for a fixed period of time in exchange for a disclosure of an invention

  4. Is every invention Patentable? • Is an improvement over a patented invention • Result is equal to or better than a result that was placed . • Result is publishable in its own right as a new scientific result. • The result solves a problem of indisputable difficulty in its field.

  5. AI & Invention • A new idea that can be logically deduced from facts that are known in a field, using transformations that are known in a field, is not considered to be inventive • Obtaining implication of given facts and rules -- Hallmark of intelligence ~ Prof. PB slides • Result: AI based on reasoning and logic cannot Invent !

  6. Genetic Algorithm • Inspired by evolutionary biology • A solution represented as a chromosome • Methodology • Initialization • Selection • Reproduction • Crossover • Mutation • Termination

  7. Genetic Programming • GP applies the approach of the genetic algorithm to the space of possible computer programs • Computer programs are the basic way for expressing the solutions to a wide variety of problems • Genetic programming now routinely delivers high-return human-competitive machine intelligence • High -> high AI (“artificial-to-intelligence” ) ratio • Routine -> repeating successfully on different set of problems • human-competitive -> is patentable in a sense

  8. GP Operators • Reproduction • Crossing over • Mutation • Architecture Alteration operation

  9. GP Flow Chart Source: John Koza slides

  10. Preparatory Steps • The human user communicates the high-level statement of the problem to the genetic programming using preparatory steps Source: www.genetic-programming.com

  11. Functional set and Terminal set • Alphabets of the programs to be made • The terminal set consists of the variables and constants of the programs • The functions are several mathematical functions and other more complex functions

  12. FITNESS MEASURE • Specifies what needs to be done • The primary mechanism for communicating the high-level statement of the problem’s requirements • The first two preparatory steps define the search space whereas the fitness measure implicitly specifies the search’s desired goal.

  13. CONTROL PARAMETERS AND TERMINATION • These steps are administrative • Control parameter: • population size. • probabilities of performing the genetic operations • the maximum size for programs • Termination criterion • maximum number of generations • may manually monitor and manually terminate • Method of designating the result • single best-so-far individual

  14. GP – Invention Machine • Problem : To create a low pass filter without patent infringement of Ladder filter. • below 1,000 Hz – Pass band • above 2,000 Hz – Stop Band • Ladder Filter Source: Genetic Programming as a Darwinian Invention Machine

  15. Program Architecture • Topology-modifying functions • alter the circuit topology • Component-creating functions • insert components into the circuit • Development-controlling functions • control the development process • Arithmetic-performing functions • specify the numerical value of the component • Automatically defined functions • enable certain substructures of the circuit to be reused

  16. Preparatory Steps • Initial Circuit • Program Architecture • Functions • Terminals • Fitness • Control Parameters • Termination

  17. Initial Circuit • Test Fixture • fixed substructure • provides access to the circuit's external input • permits probing of the circuit's output • Embryo • development occurs in the embryo Source: Genetic Programming as a Darwinian Invention Machine

  18. Functions and Terminals • F= {C, L, SERIES, PARALLEL, FLIP, TVIA0, …, TVIA7, NOOP} • Tccs = {END, CUT} • ccs- construction continuing sub-tree • END makes the modifiable component with which it is associated non-modifiable • CUT causes the component to be removed from the circuit • Taps = {R} • aps- arithmetic-performing sub-tree

  19. Fitness • Measurement the circuit’s behavior in the frequency domain • 101 Signals from 1 Hz and 100,000 Hz divided using a logarithmic scale is given • Error measured using Formula • Circuit’s similarity to the to-be-avoided ladder filter • sub graph of the given circuit that is matching to a sub graph of a ladder filter • Both are multiplied to get over all fitness • Smaller the overall value of fitness is better

  20. Control Parameters and Termination • Control Parameters • Population size, M is 1,950,000 • Circuit constructing program tree size is 300 • Termination • Goal is to generate a variety of 100%-compliant circuits • Numerous 100%-compliant circuits were harvested • Manually terminated

  21. Results • Based on Matching factor & frequency response more than 8 suitable offspring's were selected. • One of the result was elliptic filter(1927,Caur) Which is patented! Source: Genetic Programming as a Darwinian Invention Machine

  22. Conclusion • GP can automatically create design that • satisfies new specification • Avoids prior art • If a suitable fitness criteria can be found ,GP can be used in any field for invention

  23. Reference • J.R. Koza, F.H. Bennett III, and O. Stiffelman. 1999 Genetic Programming as a Darwinian Invention Machine. EuroGP’99, LNCS 1598, pp. 93-108, Ó Springer-Verlag Berlin Heidelberg 1999 • John R. Koza, Martin A. Keane, Matthew J. Streeter, "Routine High-Return Human-Competitive Evolvable Hardware," eh,pp.3, 2004 NASA/DoD Conference on Evolvable Hardware (EH'04), 2004 • http://www.genetic-programming.com • http://en.wikipedia.org/wiki/Genetic_programming • Prof. Pushpak Bhattacharyya slides

  24. Thank You

  25. Next >> Source: www.genetic-programming.com

  26. Next >> Source: www.genetic-programming.com

  27. Next >> Source: www.genetic-programming.com

More Related