Scribd
Upload
17 views

Hidden Markov Models

The document discusses hidden Markov models including their introduction, max likelihood, FB and sum-product algorithms, and Viterbi algorithm. It also covers Markov models, first and second order Markov chains, representing sequential data using latent variables, and using HMMs for speech recognition, language modeling, and analyzing biological sequences.

Uploaded by

Drij Vaghasia
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
17 views

Hidden Markov Models

The document discusses hidden Markov models including their introduction, max likelihood, FB and sum-product algorithms, and Viterbi algorithm. It also covers Markov models, first and second order Markov chains, representing sequential data using latent variables, and using HMMs for speech recognition, language modeling, and analyzing biological sequences.

Uploaded by

Drij Vaghasia
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

Hidden Markov models

• Introduction to Markov models


• Hidden Markov Models (HMM)
• Max. Likelihood
• FB algo.
• Sum-Product algo.
• Viterbi algo.
Markov models
• In these models, we assume that future predictions are independent of all but the most recent
observations.
• The simplest approach to modelling a sequence of observations is to treat them as
independent, corresponding to a graph without links. (i.i.d: independent and identically
distributed)

• However this condition is little relaxed for i.i.d., by considering a Markov model.
• The product rule can be used to express the joint distribution for a sequence of observations

• If we assume that each of the conditional distributions on the RHS is independent of all
previous observations except the most recent, we obtain the first-order Markov chain
• first order Markov model
• A first-order Markov chain of observations {xn}
in which the distribution p(xn|xn−1) of a
particular observation xn is conditioned on the
value of the previous observation xn−1.

• The joint distribution for a sequence of N


observations under this model is given by

• Similarly, second order Markov model is:


We can represent sequential data using a Markov chain of
latent variables, with each observation conditioned on the
state of the corresponding latent variable.
Hidden Markov models

• The hidden Markov model can be viewed as a specific instance of the state
space model in which the latent variables are discrete.
• The HMM is widely used in
• speech recognition
• natural language modelling
• handwriting recognition
• analysis of biological sequences such as proteins and DNA
• ....
• the latent variables are the discrete multinomial variables zn describing which component of the
mixture is responsible for generating the corresponding observation xn.

• We allow the probability distribution of zn to depend on the state of the previous latent variable zn−1
through a conditional distribution p(zn|zn−1).

Transition diagram showing a model whose latent variables have three


possible states corresponding to the three boxes.
Here, the black lines denote the elements of the transition matrix Ajk.
Emission probabilities : specification of the probabilistic model is
completed by defining the conditional distributions of the observed variables
p(xn|zn, φ), where φ is a set of parameters governing the distribution, known
as emission probabilities
Example: A weather model

• Suppose we observe gradually rising temperatures over a week: temps = [-2, 0, 2, 4, 6, 8, 10]
• We can then compute the most probable sequence of hidden states, which results into [0 0 0 0 0 1 1] telling us that
the transition from "cold" to "hot" most likely happened between the 5th and 6th days.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy