Confidence value in AdaBoost?

Ask Question

Asked 5 years, 1 month ago

Modified 5 years, 1 month ago

Viewed 278 times

I read this introduction about AdaBoost (http://www.cs.man.ac.uk/~nikolaon/~nikolaon_files/Introduction_to_AdaBoost.pdf), and am curious why confidence for each model is defined as

$$\alpha_j=\frac{1}{2}\log\frac{1-\epsilon_j}{\epsilon_j}$$

Could anyone provide any source for explaining this?

Thank you!

asked Apr 19, 2018 at 18:00

user3326682

1434 bronze badges

3

$\begingroup$ You can think AdaBoost algorithm as a example of additive modeling using exponential loss function, the confidence for each model is calculated to minimize the loss function at each step. reference: cs.cmu.edu/~epxing/Class/10701-08s/recitation/boosting.pdf $\endgroup$
– Yang Song
Apr 19, 2018 at 20:58

Add a comment |

0 Your Answer

Name

Required, but never shown

Post as a guest

Name

Required, but never shown

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.