Questions tagged [baum-welch]
The Baum–Welch algorithm is used to find the unknown parameters of a hidden Markov model (HMM).
44
questions
0
votes
0answers
7 views
baum-welch: can you show me how to derive the backwards probability?
I'm looking at the backward stage of baum-welch, for L hidden states S and L observed variables X. I don't understand how to derive this simple probability equation for $B_{L-1}(S_{L-1})$:
$P(X_{L-1}|...
1
vote
1answer
36 views
Bayes Theorem for Baum-Welch algorithm
I've been reading the Baum-Welch algorithm and somewhere it mentioned this statistical property:
$$
P(X \mid Y,Z) =\frac{ P(X,Y \mid Z)}{P(Y \mid Z)}
$$
being based on Bayes' Theorem.
I do understand ...
0
votes
0answers
14 views
Extending HMM with Gaussian Emissions to GMM
In the notation/language of HMMs, say $h_{1:T_i}^i$ be the hidden states, and $v_{1:T_i}^i$ be the observations where $i=1,\ldots,n$ denote each training set. Let each mutlivariate observation $v_t \...
1
vote
0answers
24 views
Expectation Maximisation vs Expectation Propagation in the context of Bayesian Networks
I am confused about Expectation Maximisation and Expectation Propagation algorithms in the context of Bayesian Networks, especially whether one comprise another.
What is the difference between ...
0
votes
0answers
16 views
Does this problem satisfy markov properties to be modeled as HMM?
I want to model a chemical reaction network which is defined by a stoichiometric matrix $\nu^{s\times m} $ where $s$ is the number of participating species and $m$ the number of chemical reactions. If ...
0
votes
0answers
17 views
HMM - Deal with Baum-Welch emission never observed
If I train a HMM with a given sequence of observations among n possible emissions, how do I deal with an emission that is never observed? For example, if in a 100 long observation sequence the ...
0
votes
0answers
24 views
What is an appropriate threshold for the EM algorithm?
I am implementing the Baum-Welch algorithm (special case of the EM algorithm) on a hidden Markov model and I now have to pick an appropriate stopping criteria $\epsilon$ so that the algorithm ...
0
votes
0answers
30 views
Energy based learning for HMMs: Viterbi training
I understand why we want to maximise the posterior probability to find the most likely sequence of hidden variables but I've read that this is equivalent to minimising some concept of free energy.
I'...
0
votes
0answers
45 views
HMM training of multiple observations corresponding to different hidden states
Given a set of states $\{q_1,q_2,..,q_n\}$, I am considering the following problem. Corresponding to a sequence of hidden states, I have some observations.
first sequence of hidden states ; first ...
1
vote
1answer
303 views
Example of manual implementation of baum-welch algorithm in R
Is there any code out there that implements the baum-welch algorithm for a very basic problem? It would be very helpful to actually see the algorithm in action to better understand how it works.
I ...
0
votes
0answers
17 views
Baum-Welch algorithm and link to EM algorithm [duplicate]
What is the exact link between Baum-Welch in HMM and EM?
In EM, we usually calculate:
$$Q(X) = P(X|Y,\theta)$$
and then maximize
$$\mathbb{E}_{~Q}[ log(\frac{P(Y, X, \theta)}{Q(X)})].$$
I read the ...
1
vote
0answers
44 views
Minimum Population Size for T-Test
I'm comparing two different populations with unequal variance and non normal distributions.
For sample #1 I'm drawing a random sample size of $n=30$ from a population of 200. For sample #2 I'm ...
0
votes
1answer
30 views
Deciding length of units in sound recognition for training HMMs
I am working on creating a method to detect changes from one song to another. Namely, I hope to use a Hidden Markov Model (HMM) in order to model a part of a song and check to see if it accurately ...
1
vote
1answer
346 views
Viterbi and forward-backward algorithm in HMM
I am learning HMM recently and got confused with the training problem (training model parameters and hidden state given outcome sequence).
As far as I know, both Viterbi learning and Baum-Welch (...
0
votes
0answers
71 views
Baum-Welch (EM) algorithm for non-homogeneous Hidden Markov Models
Is there a way of applying the Baum-Welch (or more general, EM) algorithm for non-homogenous Hidden Markov Models, i.e. if the Markov chain depends on covariates?
0
votes
0answers
42 views
Testing Baum Welch algorithm
I am looking for a working example of the Baum Welch algorithm of a form like this:
Define the true model $\lambda=$(A,B,$\pi$) (transition probabilities, emission probabilities, initial distribution)...
1
vote
0answers
2k views
How do I report a Welch's test?
I ran a Welch's test with Alpha= 0.05 followed by a Games Howell test. The Welch's test is not significant (p-value=0.104)
Is it enough to just mention that the groups were not significantly different ...
3
votes
1answer
578 views
What are some applications of unsupervised HMMs?
Supervised HMMs can be applied to many problems like POS tagging and OCR (optical character recognition).
I've learned that HMMs can be trained unsupervisedly using EM (Baum-Welch algorithm), what ...
2
votes
1answer
92 views
MFCCs and MoG-HMMs for speech recognition
BACKGROUND
MFCCs are coefficients which represent the most important parts of speech, and about 12 of them are used to model a one 512 points long frame (of speech). Along with them you would use ...
3
votes
2answers
4k views
Simple Explanation of Baum Welch/Viterbi
I'm looking for a very simple explanation as possible for Baum Welch and Viterbi for HMMs with a straightforward well annotated example. Almost all of the explanations I find on the net invariably ...
1
vote
0answers
137 views
What if transitions and emissions in a hidden Markov model are not independent?
A Hidden Markov Model is given by transition and emission matrices. The transition matrix determine probability of "next states" as a function of the current state. The emission matrix determine the ...
3
votes
1answer
612 views
Baum-Welch and hidden Markov models: Continuous observation densities in HMMs
I am currently trying to understand how parameter are being reestimated for hidden Markov models (HMMs), using expectation-maximization (EM).
What I seem to have problems understanding is what the ...
1
vote
1answer
53 views
Produce Multiple Sequence Alignments using Profile HMMs
I have a few multiple sequence alignments and a few groups (let's say N) of unaligned sequences. I would like to learn the emission and transition probabilities from the multiple sequence alignments, ...
3
votes
1answer
1k views
Baum Welch training of HMM
I have 200k sequences and each element of the sequence is vector of length 200. I plan to learn a HMM using this data, using the Baum-Welch EM algorithm to infer transition and emission probabilities. ...
2
votes
1answer
1k views
Forward-backward algorithm for HMM
I am currently studying this paper In which i am having some problems understanding the purpose of the forward-backwards algorithm.
First of all why even have both forward and backwards?
It seems ...
4
votes
1answer
743 views
re-estimation of emission probabilities in HMM
I am confused about the re-erstimation procedure for emissions in HMMs with Baum-Welch (still). I posted two questions concerning this general topic already and I thought I had cleared up my confusion,...
0
votes
1answer
236 views
Do re-estimated HMM parameters still need to be normalized?
A few days ago I asked this question. I only got one answer and I did not really understand it. Now I think this question is a special case of a more general question I have, namely: Do re-estiamted ...
5
votes
2answers
2k views
Scaling step in Baum-Welch algorithm
I am implementing the Baum-Welch Algorithm for training a Hidden Markov Process, to basically better understand the training process.
I have implemented the iterative procedures described in Rabiner'...
0
votes
1answer
416 views
Can the distribution of emission probabilities of an HMM be swapped out for the re-estimated ones only after all training sequences have been covered?
Regarding the re-estimation procedure of the Baum-Welch algorithm, the sources I looked into so far all describe the process in an abstract manner. Therefore I am wondering the following about ...
0
votes
1answer
120 views
combining log likelihoods from different series of observations
I have generated multiple log likelihood estimates for some parameters $\theta$ based on independent sets of observation. How can I combine these into one average?
1
vote
0answers
168 views
Hidden Markov Model EM Training: Linking states to hidden variables
I want to implement the Baum Welch Algorithm for unsupervised training of Hidden Markov Models, I am however still a little unclear on some details, especially this one:
Before learning the optimal ...
0
votes
1answer
135 views
EM algorithm update step formula
I'm using these formulas to update my initial vector, transition matrix and emission matrix:
I have update my initial vector fine using formula (1) but the other two formula have an alpha instead of ...
1
vote
1answer
366 views
Calculating alpha in EM / Baum-Welch algorithm for Hidden Markov
I am trying to use this equation
to calculate the alpha (forward) probabilities for the EM/Baum-welch algorithm but I'm running into some confusion. I don't understand what the $h_t$ is. I know its a ...
1
vote
0answers
106 views
Which parameters need to be initialized random for gaussian mixture hidden markov model?
So, if I model observation probability for a given hidden state according to a multivariate gaussian mixture model, then which parameters need to be initialized random to perform parameter re-...
1
vote
1answer
677 views
baum-welch parameter estimation numeric example
I have an HMM (picture below), with a single parameter $\theta$ I want to estimate using Baum-Welch.
I have a single training example X="HHT", and I start with an ...
2
votes
0answers
655 views
HMM (Baum-Welch) - convergence rate differences between the transition and output matrices
I am trying to learn more about the convergence properties of the Baum-Welch algorithm for estimating the HMM parameters.
I ran a test comparing the convergence of both the transition and output ...
1
vote
1answer
231 views
HMM with some known inputs
I am trying to solve a problem using HMM where the observations Yt are the result of both known and hidden parameters, but unfortunately, I am not sure how to combine the known and hidden parameters ...
1
vote
1answer
2k views
How to Train HMM model with two different sequences using the Baum-Wech algorithm
I am using HMM to visualize drinking gestures of different container types.
I began training HMM with one sequence corresponding to one container type, but I want to visualize it with python now with ...
2
votes
0answers
351 views
Baum-Welch algorithm variation for Hidden Markov model with reward
Following my previous question on the subject I would like to get your feedback on the following alternative solution.
(The original solution to this question is the usage of the POMDP model proposed ...
0
votes
1answer
249 views
Baum Welch:Calculating transition probability
I am trying to understand the Baum Welch algorithm by implementing it in xls. I have chosen a simple example of observations from a loaded (L) vs fair (F) die.
I calculate the forward and backward ...
0
votes
1answer
431 views
In what way does clustering help in classification?
I understand that with a K-means or DTW algorithm one can cluster time series using a distance criterion, i also understand that with a K-NN algorithm for example one can do pattern recognition and ...
1
vote
0answers
236 views
How do I find the hidden states of a HMM for a non time series data? How does EM Algorithm play a role in HMM?
I am a beginner in machine learning. My project requires me to model user intentions using HMM i.e based on parameters like no.of clicks, time on page, etc I have to model whether the user wants to ...
2
votes
2answers
1k views
How to model Hidden markov model with changing transition probability
I have a series of observations that fall into two outcomes, 0 or 1. These observations have an associated time of observation, as well as additional features that I can gather for that observation. I ...
18
votes
2answers
21k views
What are the differences between the Baum-Welch algorithm and Viterbi training?
I am currently using Viterbi training for an image segmentation problem. I wanted to know what the advantages/disadvantages are of using the Baum-Welch algorithm instead of Viterbi training.
