Generalization Error Bounds
One theoretical motivation behind margin classifiers is that their generalization error may be bound by parameters of the algorithm and a margin term. An example of such a bound is for the AdaBoost algorithm. Let be a set of examples sampled independently at random from a distribution . Assume the VC-dimension of the underlying base classifier is and . Then with probability we have the bound
for all .
Read more about this topic: Margin Classifier
Famous quotes containing the words error and/or bounds:
“Meanwhile, if the fear of falling into error sets up a mistrust of Science, which in the absence of such scruples gets on with the work itself, and actually cognizes something, it is hard to see why we should not turn round and mistrust this very mistrust.... What calls itself fear of error reveals itself rather as fear of the truth.”
—Georg Wilhelm Friedrich Hegel (1770–1831)
“Firmness yclept in heroes, kings and seamen,
That is, when they succeed; but greatly blamed
As obstinacy, both in men and women,
Whene’er their triumph pales, or star is tamed —
And ‘twill perplex the casuist in morality
To fix the due bounds of this dangerous quality.”
—George Gordon Noel Byron (1788–1824)