DISPERSION EFFECT ON GENERALISATION ERROR IN CLASSIFICATION - Experimental Proof and Practical Algorithm

Benoît Gandar, Gaëlle Loosli, Guillaume Deffuant

Abstract

Recent theoretical work proposes criteria of dispersion to generate learning points. The aim of this paper is to convince the reader, with experimental proofs, that dispersion is a good criterion in practice for generating learning points for classification problems. Problem of generating learning points consists then in generating points with the lowest dispersion. As a consequence, we present low dispersion algorithms existing in the literature, analyze them and propose a new algorithm.

References

  1. Gandar, B., Loosli, G., and Deffuant, G. (2009). How to optimize sample in active learning : Dispersion, an optimum criterion for classification ? In European conference ENBIS European Network for Business and Industrial Statistics.
  2. Johnson, M., Moore, L., and Ylvisaker, D. (1990). Mimimax and maximin distance designs. Journal of Statistical Planning Inference, 26(2):131-148.
  3. Lindemann, S. and LaValle, S. (2004). Incrementally Reducing Dispersion by Increasing Voronoi Bias in RRTs. In IEEE International Conference on Robotics and Automation.
  4. Niederreiter, H. (1992). Random Number Generation and Quasi-Monte Carlo Methods. Society for Industrial and Applied Mathematics.
  5. Sergent, M., Phan Tan Luu, R., and Elguero, J. (1997). Statistical Analysis of Solvent Scales. Anales de Quimica, 93(Part. 1):3-6.
  6. Teytaud, O., Gelly, S., and Mary, J. (2007). Active learning in regression, with application to stochastic dynamic programming. In Proceedings of International Conference on Informatics to Control, Automation and Robotics.
Download


Paper Citation


in Harvard Style

Gandar B., Loosli G. and Deffuant G. (2011). DISPERSION EFFECT ON GENERALISATION ERROR IN CLASSIFICATION - Experimental Proof and Practical Algorithm . In Proceedings of the 3rd International Conference on Agents and Artificial Intelligence - Volume 1: ICAART, ISBN 978-989-8425-40-9, pages 703-706. DOI: 10.5220/0003293007030706


in Bibtex Style

@conference{icaart11,
author={Benoît Gandar and Gaëlle Loosli and Guillaume Deffuant},
title={DISPERSION EFFECT ON GENERALISATION ERROR IN CLASSIFICATION - Experimental Proof and Practical Algorithm},
booktitle={Proceedings of the 3rd International Conference on Agents and Artificial Intelligence - Volume 1: ICAART,},
year={2011},
pages={703-706},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003293007030706},
isbn={978-989-8425-40-9},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 3rd International Conference on Agents and Artificial Intelligence - Volume 1: ICAART,
TI - DISPERSION EFFECT ON GENERALISATION ERROR IN CLASSIFICATION - Experimental Proof and Practical Algorithm
SN - 978-989-8425-40-9
AU - Gandar B.
AU - Loosli G.
AU - Deffuant G.
PY - 2011
SP - 703
EP - 706
DO - 10.5220/0003293007030706