Margin For Boosting Algorithms
The margin for an iterative boosting algorithm given a set of examples with two classes can be defined as follows. The classifier is given an example pair where is a domain space and is the label of the example. The iterative boosting algorithm then selects a classifier at each iteration where is a space of possible classifiers that predict real values. This hypothesis is then weighted by as selected by the boosting algorithm. At iteration, The margin of an example can thus be defined as
By this definition, the margin is positive if the example is labeled correctly and negative is the example is labeled incorrectly.
This definition may be modified and is not the only way to define margin for boosting algorithms. However, there are reasons why this definition may be appealing.
Read more about this topic: Margin Classifier
Famous quotes containing the word margin:
“Then he rang the bell and ordered a ham sandwich. When the maid placed the plate on the table, he deliberately looked away but as soon as the door had shut, he grabbed the sandwich with both hands, immediately soiled his fingers and chin with the hanging margin of fat and, grunting greedily, began to much.”
—Vladimir Nabokov (1899–1977)