Support Vector Machines

Linear Separators
• Binary classification can be viewed as the task of
separating classes in feature space:
w Tx + b = 0
w Tx + b > 0
w Tx + b < 0

f(x) = sign(wTx + b)
 Linear Separators
• Which of the linear separators is optimal?
What is a good Decision Boundary?

• Many decision
boundaries! Class 2
– The Perceptron algorithm
can be used to find such a
boundary
• Are all decision
boundaries equally Class 1

good?

4
Examples of Bad Decision Boundaries

Class 2 Class 2

Class 1 Class 1

5
 Finding the Decision Boundary
• Let {x1, ..., xn} be our data set and let yi Î {1,-1} be the class
label of xi
T
For yi=1 w xi  b 1
For yi=-1 wT xi  b  1
y=1
y=1
So:
yi w xi  b 1, xi , yi 
y=1 T
y=-1 y=1
y=-1 y=1
Class 2
y=-1
y=-1
y=-1

Class 1 y=-1
m

6
 Large-margin Decision Boundary
• The decision boundary should be as far away
from the data of both classes as possible
– We should maximize the margin, m

Class 2

Class 1
m

7
 Finding the Decision Boundary
• The decision boundary should classify all points correctly Þ

• The decision boundary can be found by solving the following

constrained optimization problem

• This is a constrained optimization problem. Solving it requires

to use Lagrange multipliers
8
 Finding the Decision Boundary

• The Lagrangian is

– ai≥0
– Note that ||w||2 = wTw
9
 Gradient with respect to w and b
• Setting the gradient of w.r.t. w and b to
zero, we have
n
1 T
L  w w    i 1  yi wT xi  b  
2 i 1

1 m k k n   m k k 
  w w    i  1  yi   w xi  b  
2 k 1 i 1   k 1 
n: no of examples, m: dimension of the space

 L
 wk 0, k

 L
 b 0
 10

 The Dual Problem
• If we substitute to , we have

Since

• This is a function of ai only

11
 The Dual Problem
• The new objective function is in terms of ai only
• It is known as the dual problem: if we know w, we know all ai; if we know
all ai, we know w
• The original problem is known as the primal problem
• The objective function of the dual problem needs to be maximized (comes
out from the KKT theory)
• The dual problem is therefore:

Properties of ai when we The result when we differentiate

introduce the Lagrange the original Lagrangian w.r.t. b
multipliers 12
 The Dual Problem

• This is a quadratic programming (QP) problem

– A global maximum of ai can always be found

• w can be recovered by

13
 Characteristics of the Solution
• Many of the ai are zero
– w is a linear combination of a small number of data
points
– This “sparse” representation can be viewed as data
compression as in the construction of knn classifier

• xi with non-zero ai are called support vectors (SV)

– The decision boundary is determined only by the SV
– Let tj (j=1, ..., s) be the indices of the s support vectors.
We can write

– Note: w need not be formed explicitly

14
A Geometrical Interpretation
Class 2

a8=0. a10=0
6
a7=0
a2=0
a5=0
a1=0.8
a4=0
a6=1.4
a9=0
a3=0
Class 1

15
 Characteristics of the Solution
• For testing with a new data z

– Compute
and classify z as class 1 if the sum is positive, and
class 2 otherwise

– Note: w need not be formed explicitly

16
 The Quadratic Programming Problem
• Many approaches have been proposed
– Loqo, cplex, etc. (see http://www.numerical.rl.ac.uk/qp/qp.html)
• Most are “interior-point” methods
– Start with an initial solution that can violate the constraints
– Improve this solution by optimizing the objective function and/or
reducing the amount of constraint violation
• For SVM, sequential minimal optimization (SMO) seems to be the
most popular
– A QP with two variables is trivial to solve
– Each iteration of SMO picks a pair of (ai,aj) and solve the QP with these
two variables; repeat until convergence
• In practice, we can just regard the QP solver as a “black-box”
without bothering how it works
17
 Non-linearly Separable Problems
• We allow “error” xi in classification; it is based on the output
of the discriminant function wTx+b
• xi approximates the number of misclassified samples

Class 2

Class 1
18
 Soft Margin Hyperplane
• The new conditions become

– xi are “slack variables” in optimization

– Note that xi=0 if there is no error for xi
– xi is an upper bound of the number of errors
n
1 2
• We want to minimize
w  C  i
2 i 1

• C : tradeoff parameter between error and margin

19
 The Optimization Problem
n n n
1 T
 
L  w w  C   i   i 1   i  yi wT xi  b 
2
    i i
i 1 i 1 i 1

With α and μ Lagrange multipliers, POSITIVE

L n
 n 
w j   y x i i ij 0 w   i yi xi 0
w j i 1 i 1

L
C   j   j 0
 j

L n
 yi i 0
b i 1
20
 The Dual Problem
1 n n T  n
L     i j yi y j xi x j  C   i 
2 i 1 j 1 i 1

n   n  n
   i  1   i  yi    j y j x j xi  b    
T
  i i
i 1   j 1  i 1

n
With
 y 0 C  j   j
i i
i 1

1 T 
n n n
L     i j yi y j xi x j    i
2 i 1 j 1 i 1
 The Optimization Problem
• The dual of this new constrained optimization problem is

C  j   j
• New constrainsderive from since μ and α are
positive.
• w is recovered as

• This is very similar to the optimization problem in the linear

separable case, except that there is an upper bound C on ai now
• Once again, a QP solver can be used to find ai

22
 n
1 2
w  C  i
2 i 1

• The algorithm try to keep ξ null, maximising the

margin
• The algorithm does not minimise the number of
error. Instead, it minimises the sum of distances fron
the hyperplane

• When C increases the number of errors tend to

lower. At the limit of C tending to infinite, the
solution tend to that given by the hard margin
formulation, with 0 errors
12/10/2024 23
Soft margin is more robust

24
Extension to Non-linear Decision Boundary
• So far, we have only considered large-margin classifier with a linear
decision boundary
• How to generalize it to become nonlinear?
• Key idea: transform xi to a higher dimensional space to “make life
easier”
– Input space: the space the point xi are located
– Feature space: the space of f(xi) after transformation
• Why transform?
– Linear operation in the feature space is equivalent to non-linear operation
in input space
– Classification can become easier with a proper transformation. In the XOR
problem, for example, adding a new feature of x1x2 make the problem
linearly separable
25
 X Y
XORIs not linearly separable
0 0 0
0 1 1
1 0 1
1 1 0

Is linearly separable
X Y XY
0 0 0 0
0 1 0 1
1 0 0 1
1 1 1 0
26
Find a feature space

27
 Transforming the Data
f( )
f( ) f( )
f( ) f( ) f( )
f(.) f( )
f( ) f( )
f( ) f( )
f( ) f( )
f( ) f( ) f( )
f( )
f( )

Input space Feature space

Note: feature space is of higher dimension
than the input space in practice

• Computation in the feature space can be costly

because it is high dimensional
– The feature space is typically infinite-dimensional!
• The kernel trick comes to rescue
28
 Transforming the Data
f( )
f( ) f( )
f( ) f( ) f( )
f(.) f( )
f( ) f( )
f( ) f( )
f( ) f( )
f( ) f( ) f( )
f( )
f( )

Input space Feature space

Note: feature space is of higher dimension
than the input space in practice

• Computation in the feature space can be costly

because it is high dimensional
– The feature space is typically infinite-dimensional!
• The kernel trick comes to rescue
29
 The Kernel Trick
• Recall the SVM optimization problem

• The data points only appear as inner product

• As long as we can calculate the inner product in the feature
space, we do not need the mapping explicitly
• Many common geometric operations (angles, distances) can
be expressed by inner products
• Define the kernel function K by
30
 An Example for f(.) and K(.,.)
• Suppose f(.) is given as follows

• An inner product in the feature space is

• So, if we define the kernel function as follows, there is no need to

carry out f(.) explicitly

• This use of kernel function to avoid carrying out f(.) explicitly is

known as the kernel trick
31
 Kernels
• Given a mapping: x  φ(x)
a kernel is represented as the inner product
K (x, y )   φ (x)φ (y)
i
i i

A kernel must satisfy the Mercer’s condition:

g (x) such that


g 2 (x)dx 0 

K (x, y ) g (x) g (y )dxdy 0

32
 Modification Due to Kernel Function
• Change all inner products to kernel functions
• For training,

Original

With
kernel
function

33
 Modification Due to Kernel Function
• For testing, the new data z is classified as class
1 if f ³0, and as class 2 if f <0

Original

With
kernel
function

34
 More on Kernel Functions
• Since the training of SVM only requires the value of
K(xi, xj), there is no restriction of the form of xi and xj
– xi can be a sequence or a tree, instead of a feature vector

• K(xi, xj) is just a similarity measure comparing xi and xj

• For a test object z, the discriminant function essentially

is a weighted sum of the similarity between z and a
pre-selected set of objects (the support vectors)

35
 Example
• Suppose we have 5 1D data points
– x1=1, x2=2, x3=4, x4=5, x5=6, with 1, 2, 6 as class 1
and 4, 5 as class 2  y1=1, y2=1, y3=-1, y4=-1, y5=1

36
 Example

class 1 class 2 class 1

1 2 4 5 6

37
 Example
• We use the polynomial kernel of degree 2
– K(x,y) = (xy+1)2
– C is set to 100

• We first find ai (i=1, …, 5) by

38
 Example
• By using a QP solver, we get
– a1=0, a2=2.5, a3=0, a4=7.333, a5=4.833
– Note that the constraints are indeed satisfied
– The support vectors are {x2=2, x4=5, x5=6}
• The discriminant function is

• b is recovered by solving f(2)=1 or by f(5)=-1 or by f(6)=1,

• All three give b=9

39
 Example
Value of discriminant function

class 1 class 2 class 1

1 2 4 5 6

40
 Kernel Functions
• In practical use of SVM, the user specifies the kernel
function; the transformation f(.) is not explicitly stated
• Given a kernel function K(xi, xj), the transformation
f(.) is given by its eigenfunctions (a concept in
functional analysis)
– Eigenfunctions can be difficult to construct explicitly
– This is why people only specify the kernel function without
worrying about the exact transformation
• Another view: kernel function, being an inner product,
is really a similarity measure between the objects

41
 A kernel is associated to a transformation
– Given a kernel, in principle it should be recovered the
transformation in the feature space that originates it.

– K(x,y) = (xy+1)2= x2y2+2xy+1

 x2 
 
It corresponds the transformation x   2x 
 1 
 
12/10/2024 42
 Examples of Kernel Functions
• Polynomial kernel up to degree d

• Polynomial kernel up to degree d

• Radial basis function kernel with width s

– The feature space is infinite-dimensional

• Sigmoid with parameter k and q

– It does not satisfy the Mercer condition on all k and q

43
Example

44
 Building new kernels
• If k1(x,y) and k2(x,y) are two valid kernels then the
following kernels are valid
– Linear Combination
k ( x, y ) c1k1 ( x, y )  c2 k 2 ( x, y )

k ( x, y ) expk1 ( x, y )
– Exponential

– Product
k ( x, y ) k1 ( x, y ) k 2 ( x, y )
– Polymomial tranfsormation (Q: polymonial with non negative
coeffients) 1
k ( x, y ) Q k ( x, y ) 
– Function product (f: any function)
k ( x, y )  f ( x ) k ( x, y ) f ( y )
1
45
 Ploynomial kernel

Ben-Hur et al, PLOS computational Biology 4 (2008)

46
 Gaussian RBF kernel

Ben-Hur et al, PLOS computational Biology 4 (2008)

47
 Spectral kernel for sequences
• Given a DNA sequence x we can count the number of
bases (4-D feature space)
1 ( x) (n A , nC , nG , nT )

• Or the number of dimers (16-D space)

2 ( x) (n AA , n AC , n AG , n AT , nCA , nCC , nCG , nCT ,..)

• Or l-mers (4l –D space)

• The spectral kernel is kl ( x, y ) l x l y 

12/10/2024 48
 Choosing the Kernel Function
• Probably the most tricky part of using SVM.
• The kernel function is important because it creates the kernel matrix,
which summarizes all the data
• Many principles have been proposed (diffusion kernel, Fisher kernel,
string kernel, …)
• There is even research to estimate the kernel matrix from available
information

• In practice, a low degree polynomial kernel or RBF kernel with a

reasonable width is a good initial try
• Note that SVM with RBF kernel is closely related to RBF neural
networks, with the centers of the radial basis functions automatically
chosen for SVM
49
 Other Aspects of SVM
• How to use SVM for multi-class classification?
– One can change the QP formulation to become multi-class
– More often, multiple binary classifiers are combined
• See DHS 5.2.2 for some discussion
– One can train multiple one-versus-all classifiers, or combine
multiple pairwise classifiers “intelligently”
• How to interpret the SVM discriminant function value as
probability?
– By performing logistic regression on the SVM output of a set of
data (validation set) that is not used for training
• Some SVM software (like libsvm) have these features built-
in
50
 Active Support Vector Learning

P. Mitra, B. Uma Shankar and S. K. Pal, Segmentation of multispectral remote sensing

Images using active support vector machines, Pattern Recognition Letters, 2004.
Supervised Classification
 Software
• A list of SVM implementation can be found at
http://www.kernel-machines.org/software.htm
l
• Some implementation (such as LIBSVM) can
handle multi-class classification
• SVMLight is among one of the earliest
implementation of SVM
• Several Matlab toolboxes for SVM are also
available
53
 Summary: Steps for Classification
• Prepare the pattern matrix
• Select the kernel function to use
• Select the parameter of the kernel function and the
value of C
– You can use the values suggested by the SVM software, or
you can set apart a validation set to determine the values
of the parameter
• Execute the training algorithm and obtain the ai
• Unseen data can be classified using the ai and the
support vectors
54
Strengths and Weaknesses of SVM
• Strengths
– Training is relatively easy
• No local optimal, unlike in neural networks
– It scales relatively well to high dimensional data
– Tradeoff between classifier complexity and error can
be controlled explicitly
– Non-traditional data like strings and trees can be
used as input to SVM, instead of feature vectors
• Weaknesses
– Need to choose a “good” kernel function.
55
 Conclusion
• SVM is a useful alternative to neural networks
• Two key concepts of SVM: maximize the
margin and the kernel trick
• Many SVM implementations are available on
the web for you to try on your data set!

56
 Resources
• http://www.kernel-machines.org/
• http://www.support-vector.net/
• http://www.support-vector.net/icml-tutorial.p
df
• http://www.kernel-machines.org/papers/tuto
rial-nips.ps.gz
• http://www.clopinet.com/isabelle/Projects/SV
M/applist.html

57