Overfitting problem and the over-training in the era of data: Particularly for Artificial Neural Networks
Authors:
- 1 December 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 173-177
Abstract
When we try to classify a set of data or to create a model to a cloud of points, different techniques can be used. Among them, Artificial Neural Networks are nowadays reinvented with the peak of the Machine Learning, Big Data, etc. In the process to find the best classification and be sure on it, one of the biggest concerns that we can come up against is the problem of overfitting. In this paper, we analyze it and set out a case study.Keywords
This publication has 15 references indexed in Scilit:
- Application of RBF neural networks and unscented transformation in probabilistic power-flow of microgrids including correlated wind/PV units and plug-in hybrid electric vehiclesSimulation Modelling Practice and Theory, 2017
- Multi-label classification via multi-target regression on data streamsMachine Learning, 2016
- Editing training data for multi-label classification with the k-nearest neighbor rulePattern Analysis and Applications, 2015
- Multi-label classification with Bayesian network-based chain classifiersPattern Recognition Letters, 2014
- An Extensive Evaluation of Decision Tree–Based Hierarchical Multilabel Classification Methods and Performance MeasuresComputational Intelligence, 2013
- A Review on Multi-Label Learning AlgorithmsIEEE Transactions on Knowledge and Data Engineering, 2013
- Tree ensembles for predicting structured outputsPattern Recognition, 2013
- Learning generative visual models from few training examples: An incremental Bayesian approach tested on 101 object categoriesComputer Vision and Image Understanding, 2007
- Theory of Classification: a Survey of Some Recent AdvancesESAIM: Probability and Statistics, 2005
- Knowledge Discovery in Multi-label Phenotype Data2001