Abstract
We study a simple learning algorithm for binary classification. Instead of predicting with the best hypothesis in the hypothesis class, that is, the hypothesis that minimizes the training error, our algorithm predicts with a weighted average of all hypotheses, weighted exponentially with respect to their training error. We show that the prediction of this algorithm is much more stable than the prediction of an algorithm that predicts with the best hypothesis. By allowing the algorithm to abstain from predicting on some examples, we show that the predictions it makes when it does not abstain are very reliable. Finally, we show that the probability that the algorithm abstains is comparable to the generalization error of the best hypothesis in the class.
Original language | English (US) |
---|---|
Pages (from-to) | 1698-1722 |
Number of pages | 25 |
Journal | Annals of Statistics |
Volume | 32 |
Issue number | 4 |
DOIs | |
State | Published - Aug 2004 |
All Science Journal Classification (ASJC) codes
- Statistics and Probability
- Statistics, Probability and Uncertainty
Keywords
- Averaging
- Bayesian methods
- Classification
- Ensemble methods
- Generalization bounds