Expectation-Maximization (EM) Algorithm & Monte Carlo Sampling for Inference and Approximation.
-
Upload
elinor-bruce -
Category
Documents
-
view
225 -
download
0
Transcript of Expectation-Maximization (EM) Algorithm & Monte Carlo Sampling for Inference and Approximation.
![Page 1: Expectation-Maximization (EM) Algorithm & Monte Carlo Sampling for Inference and Approximation.](https://reader034.fdocuments.us/reader034/viewer/2022051019/5697c0091a28abf838cc6f69/html5/thumbnails/1.jpg)
Expectation-Maximization (EM) Algorithm &Monte Carlo Sampling for Inference and Approximation
![Page 2: Expectation-Maximization (EM) Algorithm & Monte Carlo Sampling for Inference and Approximation.](https://reader034.fdocuments.us/reader034/viewer/2022051019/5697c0091a28abf838cc6f69/html5/thumbnails/2.jpg)
Expectation-Maximization Algorithm
“The Expectation-Maximization algorithm is a general technique for finding maximum likelyhood* solutions for probabilistic models having latent variables” (Dempster et al., 1977; McLachlan and Krishnan,
1997).
Is an iterative process and consists of two steps: E-step and M-step.
General purpose technique:
- Needs to be adapted for each application- Versatile. Used in machine learning, computer vision, language processing....
![Page 3: Expectation-Maximization (EM) Algorithm & Monte Carlo Sampling for Inference and Approximation.](https://reader034.fdocuments.us/reader034/viewer/2022051019/5697c0091a28abf838cc6f69/html5/thumbnails/3.jpg)
Intro: Maximum Likelihood Estimation methods
Maximum Likelihood Estimation (MLE) are methods to estimate parameters of an unknown, parameter-dependent probability density function p( x | θ ) from the observed sample (x1,x2,...,xn).
![Page 4: Expectation-Maximization (EM) Algorithm & Monte Carlo Sampling for Inference and Approximation.](https://reader034.fdocuments.us/reader034/viewer/2022051019/5697c0091a28abf838cc6f69/html5/thumbnails/4.jpg)
- When is EM useful?
- When MLE solutions are difficult or not possible to get because there are latent variables involved.
- Either missing values or we decide to get aditional unkown variables for modelling simplicity.
![Page 5: Expectation-Maximization (EM) Algorithm & Monte Carlo Sampling for Inference and Approximation.](https://reader034.fdocuments.us/reader034/viewer/2022051019/5697c0091a28abf838cc6f69/html5/thumbnails/5.jpg)
EM summarized
- Given a joint distribution p(Z , X | θ
) over obsereved variables X and latent variables Z, governed by parameters θ, the goal is to
optimize the likelihood function p( X | θ
) with respect to θ.
- Choose an inital setting for the parameters E step.
- Evaluate p(Z | X, )
- M step. Evaluate given by = arg max Q(θ, )
- Where Q(θ, )=Σ p( Z | X, ) ln p( X,Z | θ
)
- Check for convergence. If not satisfied, then ←
-from (Christopher M. Bishop, Patter Recognition and Machine Learning. Springer, 2006)
![Page 6: Expectation-Maximization (EM) Algorithm & Monte Carlo Sampling for Inference and Approximation.](https://reader034.fdocuments.us/reader034/viewer/2022051019/5697c0091a28abf838cc6f69/html5/thumbnails/6.jpg)
Monte Carlo Sampling for Inference and Approximation
- Inference – To draw conclusions from gathered data.
- Monte Carlo Sampling – Broad selection of computational algorithms that rely on repeated random sampling to obtain numerical results.
- For a better understanding we have prepared two very simple examples.
![Page 7: Expectation-Maximization (EM) Algorithm & Monte Carlo Sampling for Inference and Approximation.](https://reader034.fdocuments.us/reader034/viewer/2022051019/5697c0091a28abf838cc6f69/html5/thumbnails/7.jpg)
Rolling a dice
- We know that the probability of getting a 4 is:
- 1/6 (approx 17%)
- Can we obtain the same result by Monte Carlo simulation?
- More iterations give less error in the result!
![Page 8: Expectation-Maximization (EM) Algorithm & Monte Carlo Sampling for Inference and Approximation.](https://reader034.fdocuments.us/reader034/viewer/2022051019/5697c0091a28abf838cc6f69/html5/thumbnails/8.jpg)
Calculating the area of the unit circle
0 0.2 0.4 0.6 0.8 10
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
10 iterations of Monte Carlo
Ratio: 2.4
![Page 9: Expectation-Maximization (EM) Algorithm & Monte Carlo Sampling for Inference and Approximation.](https://reader034.fdocuments.us/reader034/viewer/2022051019/5697c0091a28abf838cc6f69/html5/thumbnails/9.jpg)
Calculating the area of the unit circle
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1000 iterations of Monte Carlo
Ratio: 3.04
![Page 10: Expectation-Maximization (EM) Algorithm & Monte Carlo Sampling for Inference and Approximation.](https://reader034.fdocuments.us/reader034/viewer/2022051019/5697c0091a28abf838cc6f69/html5/thumbnails/10.jpg)
Calculating the area of the unit circle
- 1 million iterations:
- Ratio: 3.1400
- 100 million iterations:
- Ratio: 3.1416
And so forth!
![Page 11: Expectation-Maximization (EM) Algorithm & Monte Carlo Sampling for Inference and Approximation.](https://reader034.fdocuments.us/reader034/viewer/2022051019/5697c0091a28abf838cc6f69/html5/thumbnails/11.jpg)
Application of EM
- Pattern Recognition- Image Recognition - Computer vision- Maximum likelihood- Bioinformatics
![Page 12: Expectation-Maximization (EM) Algorithm & Monte Carlo Sampling for Inference and Approximation.](https://reader034.fdocuments.us/reader034/viewer/2022051019/5697c0091a28abf838cc6f69/html5/thumbnails/12.jpg)
Application of MC
- Finance- Statistics - Molecular dynamics- Computer Graphics- Fluid mechanics