Professional Documents
Culture Documents
SVM Incremental Learning, Adaptation and Optimization - IJCNN 2003 Presentation
SVM Incremental Learning, Adaptation and Optimization - IJCNN 2003 Presentation
and Optimization
Chris Diehl Gert Cauwenberghs
Applied Physics Laboratory ECE Department
Johns Hopkins University Johns Hopkins University
Chris.Diehl@jhuapl.edu gert@jhu.edu
( )
N
Qij = yi y j K xi , x j θ f ( x ) = ∑ yiα i K ( xi , x θ ) + b
i =1
l Model Selection
Ø Repeatedly select (C,θ )
Ø Solve the quadratic program
Ø Perform cross-validation (approximate or exact)
Karush-Kuhn-Tucker (KKT) Conditions
> 0 ∀i ∈ R
∂W
gi = = yi f ( xi ) − 1 = = 0 ∀i ∈ S
∂α i < 0 ∀i ∈ E
∂W N
h= = ∑ y jα j = 0
∂b j =1
f ( x) = 1
f ( x) = − 1
Incrementing Additional Examples into the
Solution
( )
f ( x ) = ∑ yiα i K ( xi , x θ ) + ∑ y jα j K x j , x θ + b
i∈E j∈S
(
f '( x) = ∑ yiα i K ( xi , x θ ) + ∑ y j (α j + ∆α j ) K xj , x θ +
i∈E j∈S
)
∑ y ∆α K ( x , x θ ) + b + ∆b
k ∈U
k k k
U : Unlearned Vectors
(0 ≤ α < C )
Strategy
Preserve the KKT conditions y =1
for the examples in R, S and
E while incrementing the y =1
examples in U into the y =1
solution f (x) = 1
f ( x) = − 1
Part 1: Computing Coefficient and Margin
Sensitivities
∆α k ∆b
l The coefficient sensitivities β k = ∀k ∈ S and β =
∆p ∆p
will be computed by enforcing the constraints
∆g i ∆h
γ i = = 0 ∀i ∈ S and =0
∆p ∆p
l Margin sensitivities {γ i ∀i ∈ E , R, U } can then be
computed based on the coefficient sensitivities
Part 2: Bookkeeping
l Fundamental Assumption
Ø No examples change category (e.g. margin to error vector)
during a perturbation
l Perturbation Process
Ø Using the coefficient and margin sensitivities, compute the
smallest ∆p that leads to a category change
− Margin vectors: track changes in α
− Error, reserve, unlearned vectors: track changes in g
Ø Update the example categories and classifier parameters
Ø Recompute the coefficient and margin sensitivities and
repeat the process until p = 1.
Decremental Learning and Cross-Validation