Winnow (algorithm): Difference between revisions

Revision as of 21:36, 15 November 2016

Revision as of 11:33, 10 February 2017

Line 20:

* If an example is predicted incorrectly and the correct result was 1, for each feature <math>x_{i}=1</math>, the corresponding weight <math>w_{i}</math> is multiplied by <math>\alpha</math> (promotion step).

<math>\forall x_{i} = 1, w_{i} = 2w_\alpha w_{i}</math>

A typical value for