1. 1 M =3 1 M =9t t 0 01 1 0 x 1 0 x 1

2. y=1 y=0 y = 1 margin

3.

4.

5.

6.

7. A General Boosting Procedure 1. Initialize 2. From t = 1 to T. (T: ) - : Learn a weak learner - : Calculate error - : Create new distribution 3. : Combine weak learners

8. A General Boosting Procedure 1. Initialize 2. From t = 1 to T. (T: ) - : Learn a weak learner - : Calculate error - : Create new distribution 3. : Combine weak learners

9. AdaBoost Algorithm Input: 1. : Calculate initial distribution 2. From t = 1 to T. - , :Learn and Calculate Error - if then break - Calculate Weight and next distribution Outputs:

10. AdaBoost Algorithm Input: 1. : Calculate initial distribution 2. From t = 1 to T. - , :Learn and Calculate Error - if then break - Calculate Weight and next distribution Outputs:

11.

12.

13.

14.

15.

16. AdaBoost Algorithm Input: 1. : Calculate initial distribution 2. From t = 1 to T. - , :Learn and Calculate Error - if then break - Calculate Weight and next distribution Outputs:

17.

18.

19. A General Boosting Procedure 1. Initialize 2. From t = 1 to T. (T: ) - : Learn a weak learner - : Calculate error - : Create new distribution 3. : Combine weak learners

20.

21. AdaBoost Algorithm Input: 1. : Calculate initial distribution 2. From t = 1 to T. - , :Learn and Calculate Error - if then break - Calculate Weight and next distribution Outputs:

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.