Download P(x | i )

Document related concepts

Central pattern generator wikipedia , lookup

Pattern language wikipedia , lookup

Pattern recognition wikipedia , lookup

Transcript
Pattern
Classification
All materials in these slides were taken from
Pattern Classification (2nd ed) by R. O.
Duda, P. E. Hart and D. G. Stork, John Wiley
& Sons, 2000
with the permission of the authors and the
publisher
Bayes decision theory
febr. 17.
2
Classification
Supervised learning: Based on training examples (E),
learn a modell which works fine on previously unseen
examples.
Classification: a supervised learning task of categorisation
of entities into predefined set of classes
3
Pattern Classification, Chapter 2 (Part 1)
4
Posterior, likelihood, evidence
P(j | x) = P(x | j) . P (j) / P(x)
Posterior = (Likelihood. Prior) / Evidence
Where in case of two categories
j2
P ( x )   P ( x |  j )P (  j )
j 1
Pattern Classification, Chapter 2 (Part 1)
5
Pattern Classification, Chapter 2 (Part 1)
6
Bayesian Decision
•
Decision given the posterior probabilities
X is an observation for which:
if P(1 | x) > P(2 | x)
if P(1 | x) < P(2 | x)
True state of nature = 1
True state of nature = 2
This rule minimizes the probability of the error.
Pattern Classification, Chapter 2 (Part 1)
Bayesian Decision Theory –
Generalization
7
• Use of more than one feature
• Use more than two states of nature
• Allowing actions and not only decide on the state of
•
nature
Introduce a loss of function which is more general than
the probability of error
Pattern Classification, Chapter 2 (Part 1)
8
Let {1, 2,…, c} be the set of c states of nature
(or “categories”)
Let {1, 2,…, a} be the set of possible actions
Let (i | j) be the loss incurred for taking
action i when the state of nature is j
Pattern Classification, Chapter 2 (Part 1)
Bayes decision theory example
Automatic trading (on stock exchanges)
1: the prices will increase (in the future!)
2: the prices will be lower
3: the prices won’t change too much
We cannot observe  (latent)!
1: buy
2: sell
x: actual prices (and historical prices)
x is observed
: how much to lose with an action
10
Overall risk
R = Sum of all R(i | x) for i = 1,…,a
Conditional risk
Minimizing R
Minimizing R(i | x) for i = 1,…, a
j c
R(  i | x )    (  i |  j )P (  j | x )
j 1
for i = 1,…,a
Pattern Classification, Chapter 2 (Part 1)
11
Select the action i for which R(i | x) is minimum
R is minimum and R in this case is called the
Bayes risk = best performance that can be achieved!
Pattern Classification, Chapter 2 (Part 1)
12
• Two-category classification
1 : deciding 1
2 : deciding 2
ij = (i | j)
loss incurred for deciding i when the true state of nature is j
Conditional risk:
R(1 | x) = 11P(1 | x) + 12P(2 | x)
R(2 | x) = 21P(1 | x) + 22P(2 | x)
Pattern Classification, Chapter 2 (Part 1)
13
Our rule is the following:
if R(1 | x) < R(2 | x)
action 1: “decide 1” is taken
This results in the equivalent rule :
decide 1 if:
(21- 11) P(x | 1) P(1) >
(12- 22) P(x | 2) P(2)
and decide 2 otherwise
Pattern Classification, Chapter 2 (Part 1)
14
Likelihood ratio:
The preceding rule is equivalent to the following rule:
P ( x |  1 ) 12   22 P (  2 )
if

.
P ( x |  2 )  21  11 P (  1 )
Then take action 1 (decide 1)
Otherwise take action 2 (decide 2)
Pattern Classification, Chapter 2 (Part 1)
15
Exercise
Select the optimal decision where:
= {1, 2}
P(x | 1)
P(x | 2)
P(1) = 2/3
P(2) = 1/3
N(2, 0.5) (Normal distribution)
N(1.5, 0.2)
1 2


3
4


Pattern Classification, Chapter 2 (Part 1)
Zero-one loss function
(Bayes Classifier)
0 i  j
 (  i , j )  
1 i  j
16
i , j  1 ,..., c
Therefore, the conditional risk is:
j c
R(  i | x )    (  i |  j )P (  j | x )
j 1
  P(  j | x )  1  P(  i | x )
j 1
“The risk corresponding to this loss function is the
average probability error”
Pattern Classification, Chapter 2 (Part 2)
Classifiers, Discriminant Functions
and Decision Surfaces
17
• The multi-category case
• Set of discriminant functions gi(x), i = 1,…, c
• The classifier assigns a feature vector x to class i
if:
gi(x) > gj(x) j  i
Pattern Classification, Chapter 2 (Part 2)
18
• Let gi(x) = - R(i | x)
(max. discriminant corresponds to min. risk!)
• For the minimum error rate, we take
gi(x) = P(i | x)
(max. discrimination corresponds to max.
posterior!)
gi(x)  P(x | i) P(i)
gi(x) = ln P(x | i) + ln P(i)
(ln: natural logarithm!)
Pattern Classification, Chapter 2 (Part 2)
19
• Feature space divided into c decision regions
if gi(x) > gj(x) j  i then x is in Ri
(Ri means assign x to i)
• The two-category case
• A classifier is a “dichotomizer” that has two discriminant
functions g1 and g2
Let g(x)  g1(x) – g2(x)
Decide 1 if g(x) > 0 ; Otherwise decide 2
Pattern Classification, Chapter 2 (Part 2)
20
• The computation of g(x)
g( x )  P (  1 | x )  P (  2 | x )
P( x | 1 )
P( 1 )
 ln
 ln
P( x |  2 )
P(  2 )
Pattern Classification, Chapter 2 (Part 2)
21
Discriminant functions
of the Bayes Classifier
with Normal Density
Pattern Classification, Chapter 2 (Part 1)
22
•
The Normal Density
Univariate density
•
•
•
•
Density which is analytically tractable
Continuous density
A lot of processes are asymptotically Gaussian
Handwritten characters, speech sounds are ideal or prototype
corrupted by random process (central limit theorem)
P( x ) 
2

1
1 x 
exp   
 ,
2 
 2    
Where:
 = mean (or expected value) of x
2 = expected squared deviation or variance
Pattern Classification, Chapter 2 (Part 2)
23
Pattern Classification, Chapter 2 (Part 2)
24
•
Multivariate density
•
Multivariate normal density in d dimensions is:
P( x ) 
1
( 2 )
d/2

1/ 2
 1

t
1
exp  ( x   )  ( x   )
 2

where:
x = (x1, x2, …, xd)t (t stands for the transpose vector form)
 = (1, 2, …, d)t mean vector
 = d*d covariance matrix
|| and -1 are determinant and inverse respectively
Pattern Classification, Chapter 2 (Part 2)
Discriminant Functions for the Normal
Density
25
• We saw that the minimum error-rate classification
can be achieved by the discriminant function
gi(x) = ln P(x | i) + ln P(i)
• Case of multivariate normal
1
1
d
1
t
g i ( x )   ( x   i )  ( x   i )  ln 2  ln  i  ln P (  i )
2
2
2
i
Pattern Classification, Chapter 2 (Part 3)
26
• Case i = 2.I
(I stands for the identity matrix)
g i ( x )  w it x  w i 0 (linear discriminant function)
where :
i
1
t
wi  2 ; wi 0  

i  i  ln P (  i )
2

2
(  i 0 is called the threshold for the ith category! )
Pattern Classification, Chapter 2 (Part 3)
27
• A classifier that uses linear discriminant functions
is called “a linear machine”
• The decision surfaces for a linear machine are
pieces of hyperplanes defined by:
gi(x) = gj(x)
Pattern Classification, Chapter 2 (Part 3)
28
The hyperplane is always orthogonal to the line linking the means!
Pattern Classification, Chapter 2 (Part 3)
29
• The hyperplane separating Ri and Rj
1
2
x0  (  i   j ) 
2
i   j
2
P(  i )
ln
( i   j )
P(  j )
always orthogonal to the line linking the means!
1
if P (  i )  P (  j ) then x0  (  i   j )
2
Pattern Classification, Chapter 2 (Part 3)
30
Pattern Classification, Chapter 2 (Part 3)
31
Pattern Classification, Chapter 2 (Part 3)
32
• Case i =  (covariance of all classes are
identical but arbitrary!)
• Hyperplane separating Ri and Rj


ln P (  i ) / P (  j )
1
x0  (  i   j ) 
.(  i   j )
t
1
2
( i   j )  ( i   j )
(the hyperplane separating Ri and Rj is generally
not orthogonal to the line between the means!)
Pattern Classification, Chapter 2 (Part 3)
33
Pattern Classification, Chapter 2 (Part 3)
34
Pattern Classification, Chapter 2 (Part 3)
35
• Case i = arbitrary
•
The covariance matrices are different for each category
g i ( x )  x tWi x  w it x  w i 0
where :
1 1
Wi    i
2
w i   i 1  i
1 t 1
1
w i 0    i  i  i  ln  i  ln P (  i )
2
2
(Hyperquadrics which are: hyperplanes, pairs of
hyperplanes, hyperspheres, hyperellipsoids,
hyperparaboloids, hyperhyperboloids)
Pattern Classification, Chapter 2 (Part 3)
36
Pattern Classification, Chapter 2 (Part 3)
37
Pattern Classification, Chapter 2 (Part 3)
38
Exercise
Select the optimal decision where:
= {1, 2}
P(x | 1)
P(x | 2)
N(2, 0.5) (Normal distribution)
N(1.5, 0.2)
P(1) = 2/3
P(2) = 1/3
Pattern Classification, Chapter 2
39
Parameter estimation
Pattern Classification, Chapter 3
• Data availability in a Bayesian framework
• We could design an optimal classifier if we knew:
•
•
P(i) (priors)
P(x | i) (class-conditional densities)
Unfortunately, we rarely have this complete information!
• Design a classifier from a training sample
• No problem with prior estimation
• Samples are often too small for class-conditional estimation
(large dimension of feature space!)
1
• A priori information about the problem
• E.g. assume normality of P(x | i)
P(x | i) ~ N( i, i)
Characterized by 2 parameters
• Estimation techniques
• Maximum-Likelihood (ML) and the Bayesian estimations
• Results are nearly identical, but the approaches are different
1
• Parameters in ML estimation are fixed but
unknown!
• Best parameters are obtained by maximizing the
probability of obtaining the samples observed
• Bayesian methods view the parameters as
random variables having some known distribution
• In either approach, we use P(i | x)
for our classification rule!
1
• Use the information
provided by the training samples to estimate
 = (1, 2, …, c), each i (i = 1, 2, …, c) is associated with each
category
• Suppose that D contains n samples, x1, x2,…, xn
k n
P ( D |  )   P ( xk |  )
k 1
P( D |  ) is called the likelihood of  w.r.t. the set of samples)
• ML estimate of  is, by definition the value that
maximizes P(D | )
“It is the value of  that best agrees with the actually observed
training sample”
2
• Optimal estimation
•
Let  = (1, 2, …, p)t and let  be the gradient operator
 

 
  
,
,...,

 p 
 1  2
•
•
t
We define l() as the log-likelihood function
l() = ln P(D | )
New problem statement:
determine  that maximizes the log-likelihood
ˆ  argmaxl()

2
Bayesian Estimation
• In MLE  was supposed fix
• In BE  is a random variable
• The computation of posterior probabilities P(i | x)
•
lies at the heart of Bayesian classification
Goal: compute P(i | x, D)
Given the sample D, Bayes formula can be written
P(i | x, D ) 
P(x | i , D ).P(i | D )
c
 P(x |  j , D ).P( j | D )
j 1
ter
1
• Bayesian Parameter Estimation: Gaussian
Case
Goal: Estimate  using the a-posteriori density
P( | D)
• The univariate case: P( | D)
 is the only unknown parameter
P(x |  ) ~ N(,  2 )
2
P( ) ~ N( 0 ,  0 )
(0 and 0 are known!)
47
P(D |  ).P( )
P( | D ) 
 P(D |  ).P( )d
(1)
k n
   P(x k |  ).P( )
k 1
• Reproducing density
P( | D ) ~ N(n , n2 )
Identifying (1) and (2) yields:
 n 20

2
 ˆ 
 n  
. 0
2
2 n
2
2
n 0  
 n0  0   
and n2 
 20  2
n 20   2
(2)
• The univariate case P(x | D)
• P( | D) computed
• P(x | D) remains to be computed!
P(x | D )   P(x | ).P( | D )d is Gaussian
It provides:
P(x | D ) ~ N(n , 2  n2 )
(Desired class-conditional density P(x | Dj, j))
Therefore: P(x | Dj, j) together with P(j) and using Bayes
formula, we obtain the Bayesian classification rule:



Max P( j | x, D  Max P( x |  j , D j ).P( j )
j
j

• Bayesian Parameter Estimation: General
Theory
• P(x | D) computation can be applied to any
situation in which the unknown density can be
parametrized: the basic assumptions are:
• The form of P(x | ) is assumed known, but the value of 
•
•
is not known exactly
Our knowledge about  is assumed to be contained in a
known prior density P()
The rest of our knowledge  is contained in a set D of n
random variables x1, x2, …, xn that follows P(x)
5
The basic problem is:
“Compute the posterior density P( | D)”
then “Derive P(x | D)”
Using Bayes formula, we have:
P(D | ).P()
P( | D ) 
,
 P(D | ).P()d
And by independence assumption:
k n
P(D | )   P(x k | )
k 1
52
MLE vs. Bayes estimation
• If n→∞ they are equal!
• MLE
• Simple and fast (convex optimisation vs. numerical
integration)
• Bayes estimation
• We can express our
uncertainty by P()
Sumamry
• Bayes decision theory
General framework for probabilistic decision making
• Bayes classifier
Classification is a special decision making
(1 : choose 1)
• Zero-one loss function
 can be omitted
• Bayes classifier with zero-one loss
with Normal Density
Summary
• Parameter estimation
• General procedures for densities’ parameter estimation
based on a sample
(it can be applied beyond Bayes classifier)
• Bayesian Machine learning: the marrige of Bayesian
Decision Theory and Parameter estimation from a
training sample