- 1 - Digitale Signalverarbeitung und Mustererkennung
Maximum Likelihood
Parameter Estimation
- 2 - Digitale Signalverarbeitung und Mustererkennung
Motivation: Training of HMMs
Given:
Sample of feature vectors produced by an HMM state (training data)
Wanted:
Emission probability density of the state
Approach so far:
Assume normal distribution with independent components
Compute empirical mean- and variance vector from training data
- 3 - Digitale Signalverarbeitung und Mustererkennung
Simplification: numbers instead of vectors
Random variable X with unknown density function.
2, 5, 3 Sample of X:
Given:
Wanted:
Density function for X.
- 4 - Digitale Signalverarbeitung und Mustererkennung
Normal distribution with
Simplification: numbers instead of vectors
Random variable X with unknown density function.
2, 5, 3 Sample of X:
Given:
Wanted:
Density function for X.
- 5 - Digitale Signalverarbeitung und Mustererkennung
Other density function
solution is not unique
Probably a bad choice:
„Overfitting“
Simplification: numbers instead of vectors
Random variable X with unknown density function.
2, 5, 3 Sample of X:
Given:
Wanted:
Density function for X.
- 6 - Digitale Signalverarbeitung und Mustererkennung
Correct problem formulation
Random variable X with parametric density function
Sample of X:
Given:
Wanted:
Values for the parameters , such that the likelihood of the observed sample is maximum, i.e.
Normal distribution with parameters Example:
Likelihood function
- 7 - Digitale Signalverarbeitung und Mustererkennung
Find such that is maximum.
Set derivatives to zero, i.e.
Task
Problem
Derivatives of products with many factors leads to complicated terms!
Idea: take the logarithm!
Observation Method
and
have their maximum for the same value of , as ln is monotonic increasing!
Log likelihood
function
- 8 - Digitale Signalverarbeitung und Mustererkennung
Example:
Likelihood function
Log likelihood function
Maximum likelihood estimation of parameters and 2 of the normal distribution from a sample x1, x2, …, xn
- 9 - Digitale Signalverarbeitung und Mustererkennung
Log likelihood function
Partial derivative wrt.
- 10 - Digitale Signalverarbeitung und Mustererkennung
Log likelihood function
Partial derivative wrt. 2
- 11 - Digitale Signalverarbeitung und Mustererkennung
Maximum likelihood estimation of parameters and 2 of the normal distribution from a sample x1, x2, …, xn
Result:
- 12 - Digitale Signalverarbeitung und Mustererkennung
Example:
Likelihood function (assume all xi 0)
Log likelihood function
Maximum likelihood estimation of parameter of the exponential distribution from a sample x1, x2, …, xn
Density of the exponential distribution
if
otherwise
- 13 - Digitale Signalverarbeitung und Mustererkennung
Log likelihood function
Derivative wrt.
Inverse of the empirical mean
- 14 - Digitale Signalverarbeitung und Mustererkennung
Example:
Maximum likelihood estimation of parameters a,b of the uniform distribution from a sample x1, x2, …, xn
Density of the uniform distribution
Likelihood function
if
otherwise
maximum if
minimum and
if and
otherwise
- 15 - Digitale Signalverarbeitung und Mustererkennung
Example:
Likelihood funktion
Log likelihood funktion
Maximum likelihood extimation of parameter p of a Bernoulli experiment from a sample x1, x2, …, xn
Probability distribution (discrete distribution!)
- 16 - Digitale Signalverarbeitung und Mustererkennung
Log likelihood function
Derivative wrt. p