Learning with ensembles: How over-(cid:12)tting can be useful

By Peter Sollich, Anders Krogh. 1996

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Yaoyao Liu
2. Bernt Schiele
3. Qianru Sun
4. Georg Peters
5. Tie-Yan Liu
6. Pawan Lingras
7. Han-Jia Ye
8. Dominik Lzak

Citations in the corpus

• An Ensemble of Epoch-Wise Empirical Bayes for Few-Shot Learning (2019, methodology): “It is a strategy [26, 85] to use multiple algorithms to improve machine learning performance, and which is proved to be effective to reduce the problems related to overfitting [35, 68].”
• Rough Sets: Selected Methods and Applications in Management and Engineering (2012, methodology): “used rough sets in an attempt to identify and pick top stock performers [31] while Shen and Loh predicted stock market movements by retrieving knowledge that would guide investors to decide when to buy and sell [48].”
• Ensemble-Compression: A New Method for Parallel Training of Deep Neural Networks (2016, background): “In order to improve the performance of ensemble, it is necessary to generate diverse local models other than merely accurate ones [17, 24].”
• Ensembling neural networks: Many could be better than all (2002, unknown): “Neural network ensemble is a learning paradigm where a collection of a finite number of neural networks is trained for the same task [ 42 ].”
• Diversity creation methods: a survey and categorisation (2004, background): “We in fact would desire overfitting in individual learners to emerge, as this reduces the bias component of the error, leaving the variance component to be reduced in the ensemble combination [64,65].”
• Hybrid and ensemble-based soft computing techniques in bankruptcy prediction: a survey (2010, background): “of ensemble learning’’ (Sollich and Krogh 1996).”