Krzysztof Krawiec


Home

Research:

edit SideBar

We propose a method of knowledge reuse for an ensemble of genetic programming-based learners solving a visual learning task. First, we introduce a visual learning method that uses genetic programming individuals to represent hypotheses. Individuals-hypotheses process image representation composed of visual primitives derived from the training images that contain objects to be recognized. The process of recognition is generative, i.e., an individual is supposed to restore the shape of the processed object by drawing its reproduction on a separate canvas. This canonical method is extended with a knowledge reuse mechanism that allows a learner to import genetic material from hypotheses that evolved for the other decision classes (object classes). We compare the performance of the extended approach to the basic method on a real-world tasks of handwritten character recognition, and conclude that knowledge reuse leads to significant convergence speedup and, more importantly, significantly reduces the risk of overfitting.

@INPROCEEDINGS { jaskowski07reuse,
    AUTHOR = { Wojciech Ja\'skowski and Krzysztof Krawiec and Bartosz Wieloch },
    TITLE = { Knowledge Reuse in Genetic Programming Applied to Visual Learning },
    BOOKTITLE = { GECCO '07: Proceedings of the 9th annual conference on Genetic and evolutionary computation },
    YEAR = { 2007 },
    EDITOR = { Dirk Thierens et al. },
    VOLUME = { 2 },
    PAGES = { 1790--1797 },
    ADDRESS = { London },
    MONTH = { jul },
    ORGANIZATION = { ACM SIGEVO },
    PUBLISHER = { ACM Press },
    ABSTRACT = { We propose a method of knowledge reuse for an ensemble of genetic programming-based learners solving a visual learning task. First, we introduce a visual learning method that uses genetic programming individuals to represent hypotheses. Individuals-hypotheses process image representation composed of visual primitives derived from the training images that contain objects to be recognized. The process of recognition is generative, i.e., an individual is supposed to restore the shape of the processed object by drawing its reproduction on a separate canvas. This canonical method is extended with a knowledge reuse mechanism that allows a learner to import genetic material from hypotheses that evolved for the other decision classes (object classes). We compare the performance of the extended approach to the basic method on a real-world tasks of handwritten character recognition, and conclude that knowledge reuse leads to significant convergence speedup and, more importantly, significantly reduces the risk of overfitting. },
    COMMENT = { 1 },
    FILE = { jaskowski07reuse.pdf:\\jaskowski07reuse.pdf:PDF },
    URL = { http://www.cs.bham.ac.uk/~wbl/biblio/gecco2007/docs/p1790.pdf },
    ISBN = { 978-1-59593-698-1 },
    KEYWORDS = { evolutionary computation, genetic programming, genetics-based machine learning, knowledge reuse, pattern recognition, character recognition },
    ORDER = { 0 },
    TIMESTAMP = { 2007.07.07 },
}


Powered by PmWiki