본문 바로가기
  • Home

Parameter Estimation in Debris Flow Deposition Model Using Pseudo Sample Neural Network

  • Journal of The Korea Society of Computer and Information
  • Abbr : JKSCI
  • 2012, 17(11), pp.11-18
  • Publisher : The Korean Society Of Computer And Information
  • Research Area : Engineering > Computer Science

Gyeongyong Heo 1 이창우 2 Park Choong Shik 3

1동의대학교
2국립산림과학원 남부산림연구소
3영동대학교

Accredited

ABSTRACT

Debris flow deposition model is a model to predict affected areas by debris flow and randomwalk model (RWM) was used to build themodel. Although themodel was proved to be effective in the prediction of affected areas, the model has several free parameters decided experimentally. There are several well-knownmethods to estimate parameters, however, they cannot be applied directly to the debris flowproblemdue to the small size of training data. In this paper, a modified neural network, called pseudo sample neural network (PSNN), was proposed to overcome the sample size problem. In the training phase, PSNNuses pseudo samples, which are generated using the existing samples. The pseudo samples smooth the solution space and reduce the probability of falling into a local optimum. As a result, PSNNcan estimate parameter more robustly than traditional neural networks do. All of these can be proved through the experiments using artificial and real data sets.

Citation status

* References for papers published after 2022 are currently being built.