Académique Documents
Professionnel Documents
Culture Documents
to Adaboost
Hongbo Deng
6 Feb, 2007
1
Some of the slides are borrowed from Derek Hoiem & Jan Sochman.
Outline
Background
Adaboost Algorithm
Theory/Interpretations
2
Whats So Good About Adaboost
Can be used with many different classifiers
Simple to implement
3
Resampling for
A Brief History estimating statistic
Bootstrapping
4
Bootstrap Estimation
Repeatedly draw n samples from D
For each set of samples, estimate a
statistic
The bootstrap estimate is the mean of the
individual estimates
Used to estimate a statistic (parameter)
and its variance
5
Bagging - Aggregate Bootstrapping
For i = 1 .. M
Draw n*<n samples from D with replacement
Learn classifier Ci
D3 D
6
Boosting (Schapire 1989)
Consider creating three component classifiers for a two-category problem
through boosting.
Randomly select n1 < n samples from D without replacement to obtain D1
Train weak learner C1
D2 -
++ -
7
Adaboost - Adaptive Boosting
Instead of resampling, uses training set re-weighting
Each training sample uses a weight to determine the probability
of being selected for a training set.
8
Adaboost Terminology
ht(x) weak or basis classifier (Classifier =
Learner = Hypothesis)
strong or final classifier
9
Discrete Adaboost Algorithm
Each training sample has a
weight, which determines the
probability of being selected for
training the component classifier
10
Find the Weak Classifier
11
Find the Weak Classifier
12
The algorithm core
13
Reweighting
y * h(x) = 1
y * h(x) = -1
14
Reweighting
17
Algorithm recapitulation
18
Algorithm recapitulation
19
Algorithm recapitulation
20
Algorithm recapitulation
21
Algorithm recapitulation
22
Algorithm recapitulation
23
Algorithm recapitulation
24
Pros and cons of AdaBoost
Advantages
Very simple to implement
Does feature selection resulting in relatively
simple classifier
Fairly good generalization
Disadvantages
Suboptimal solution
Sensitive to noisy data and outliers
25
References
Duda, Hart, ect Pattern Classification
Jin, Liu, etc (CMU) A New Boosting Algorithm Using Input-Dependent Regularizer
Schapire, Freund, etc Boosting the Margin: A New Explanation for the Effectiveness of Voting Methods
26
Appendix
Bound on training error
Adaboost Variants
27
Bound on Training Error (Schapire)
28
Discrete Adaboost (DiscreteAB)
(Friedmans wording)
29
Discrete Adaboost (DiscreteAB)
(Freund and Schapires wording)
30
Adaboost with Confidence
Weighted Predictions (RealAB)
31
Adaboost Variants Proposed By
Friedman
LogitBoost
Solves
Requires care to avoid numerical problems
GentleBoost
Update is fm(x) = P(y=1 | x) P(y=0 | x) instead of
Bounded [0 1]
32
Adaboost Variants Proposed By
Friedman
LogitBoost
33
Adaboost Variants Proposed By
Friedman
GentleBoost
34
Thanks!!!
Any comments or questions?
35