Stack Overflow
Stack Overflow

Questions tagged [expectation-maximization]

Expectation Maximization (often abbreviated EM) is an iterative algorithm that can be used for maximum likelihood estimation in the presence of missing data or hidden variables.

125 questions
0
votes
1 answer

EM Algorithm not working

I am trying to implement a simple EM algorithm. So far, it seems to be working well except for the small problem that variances quickly shrink to zero, converging around the mean of the data. (If I do not update the variances, it will converge to…
Nabushika
  • 364
  • 1
  • 17
0
votes
1 answer

Binomial Distribution in in R

I applied this code and I get this error over and over again. library(mixtools) simulate <- function(lambda=0.3, mu=c(0, 4), sd=c(1, 1), n.obs=10^5) { x1 <- rnorm(n.obs, mu[1], sd[1]) x2 <- rnorm(n.obs, mu[2], sd[2]) …
lolo
  • 646
  • 2
  • 7
  • 19
0
votes
0 answers

expectation maximization with constriants

I have a very general question. As we all know, in Bayesian Inference, we introduce several hyper-parameters to define a prior PDF, together with the likelihood function, we can obtain a posterior PDF from which we can obtain a…
zhengqi gao
  • 13
  • 4
0
votes
1 answer

GMM/EM on time series cluster

According to a paper, it is supposed to work. But as a learner of scikit-learn package.. I do not see how. All the sample codes cluster by ellipses or circles as here. I would really like to know how to cluster the following plot by different…
dia
  • 431
  • 2
  • 7
  • 22
0
votes
1 answer

Baum Welch (EM Algorithm) likelihood (P(X)) is not monotonically converging

So I am sort of an amateur when comes to machine learning and I am trying to program the Baum Welch algorithm, which is a derivation of the EM algorithm for Hidden Markov Models. Inside my program I am testing for convergence using the probability…
I. Cantrell
  • 170
  • 1
  • 7
0
votes
1 answer

EM algo returns different answers using pykalman

I came across a problem when I was using pykalman 0.9.5 in Python 3.6.3 Refer to the code below, why are the results from kf2 and kf3 are different while the results from kf1 and kf3 are identical? The difference of process between kf2 and kf3 is…
Gabriel
  • 161
  • 2
  • 11
0
votes
1 answer

parameters in the EM algorithm

I have been learning about the EM algorithm by using the material of Andrew Ng from Stanford, the link is here: http://cs229.stanford.edu/notes/cs229-notes7b.pdf And I have been trying to understand the implementation of the EM by using a Python…
Little
  • 3,363
  • 10
  • 45
  • 74
0
votes
1 answer

How to use EM algorithms to determine parameters(eps,minpts) of DBSCAN over one dataset?

Recently I choose to use DBSCAN clustering over a public data set. But the parameters Eps and minpts are so sensitive that it's quite hard to get good parameter values with good performance over whole data set. There seems to be over-fitting when…
ZHENGZhuang
  • 5
  • 3
0
votes
1 answer

updates in mean-field variational inference

I'm wondering if ELBO should be increasing monotonically as every update of the variational parameter q_i? I understand that it should increase at every iteration, but is this also true within one iteration for each q_i?
Y. He
  • 33
  • 6
0
votes
0 answers

Expectation Maximization in C#

In the Internet, there are many code examples for C ++ or Python. But there is not a single example in C #. I tried it myself and got an error. EM em = EM.Create(); em.ClustersNumber = 2; em.CovarianceMatrixType = (int) EM.Types.CovMatGeneric; //…
T2k
  • 11
  • 3
0
votes
1 answer

EM algorithm for multivariate t mixed models

I'm trying to implement an EM algorithm for family data where I'm assuming my observations have a multivariate t distribution. I have only two siblings per family, so all of the family groups have only two observations. Basically I'm trying to…
pa_ka
  • 65
  • 1
  • 6
0
votes
1 answer

Would non-decrease property change if the denominator is a non-invertible matrix in M step of EM algorithm?

Suppose in the M-step of EM algorithm, the denominator of some parameters are matrix and they are non-invertible, we would use pseudo inverse matrix instead of it. If so, would the log likelihood still always increase? I couldn't give a specific…
Lazar
  • 11
  • 4
0
votes
0 answers

univariate Normal impution with EM algorithm

I need some example on how to fill missing data with EM algorithm. The data, as the daily relative change of a stock price, assumed to normally distributed and univariate sample. I've done some literature search but hardly found any example on this.…
VincentHall
  • 37
  • 1
  • 1
  • 6
0
votes
1 answer

Error in R code in EM Algorithm from R blogger webpage

I refer to the EM code from this page: http://www.r-bloggers.com/imputing-missing-data-with-expectation-maximization/ Initially, when I run the code raw < - replicate(10, rpois(50,100)), I received this error: Error in raw < -replicate(10,…
Jas
  • 21
  • 5
0
votes
1 answer

expectation maximization algorithm matlab out of memory error

I am implementing Expectation Maximization algorithm in matlab. Algorithm is operating on 214096 x 2 data matrix and While computing probabilities, there is multiplication of ( 214096 x 2 ) * (2 x 2) * ( 2 x 214096 ) matrices, which is resulting in…
Umar
  • 44
  • 3
1 2 3
…
8 9