Top Banner
Hidden Markov Models, Bayesian Networks Stat 430
21

Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Mar 13, 2019

Download

Documents

phungphuc
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Hidden Markov Models, Bayesian Networks

Stat 430

Page 2: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Outline

• Definition of HMM

• Set-up of 3 main problems

• Three Main algorithms:

• Forward/Backward

• Viterbi

• Baum-Welch

Page 3: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Hidden Markov Models (HMM)

• Ideaeach state of a Markov chain emits a letter from a fixed alphabetdistribution of letters is time independent, but depends on state

• Situationusually we have string of emitted symbols, but don’t know Markov Chain (it’s hidden)

Page 4: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Application Areas

• Pattern recognition

• Search algorithms

• Sequence alignments

• Time series analysis

Page 5: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Setup

Markov State Diagramwith transition probabilities A

Emitted Sequence Y

emission probabilities B

Usually, we only observe Y (several instances of it)

Page 6: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Example

• Suppose we have five amino acid sequencesCAEFTPAVHCKETTPADHCAETPDDHCAEFDDHCDAEFPDDH

• find best possible alignment of all sequences (allowing insertions and deletions)

Page 7: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Run of an HMM• 2-step process:

initial -> emission -> transition -> emission -> transition -> q1 O1 q2 O2 q3

• Sequence of visited states Q = q1 q2 q3 ...

• Sequence of emitted symbols O = O1 O2 O3 ...

• usually we can observe O, but don’t know Q

Page 8: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Example• Markov Chain with two states, S1 and S2

and transition matrix P

• Emission alphabet {a, b}

• In S1 emission probabilitiesfor a and b are 0.5 and 0.5

• In S2 emission probabilitiesfor a and b are 0.25 and 0.75

• Observed sequence is bbb

S1 S2

S1

S2

0.9 0.1

0.8 0.2

Page 9: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Example

• Observed sequence is bbb

• What is the most likely sequence Q that emitted bbb? argmaxQ P(Q|O)

• What is the probability to observe O? P(O) = ∑Q P(O|Q) P(Q)

S1 S2

S1

S2

0.9 0.10.8 0.2

a bS1

S2

0.5 0.50.25 0.75

transitions emissions

Page 10: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Definition

• set of states S1, S2, ..., SN

• the transition matrix P with pij = P(qt+1 = Sj| qt=Si)

• an alphabet of M unique, observed symbols A = {a1, ..., aM}

• emission probabilities bi(a) = P(state Si emits a)

• initial distribution πi = P(q1 = Si)

A hidden Markov model (HMM) consists of

Page 11: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Three Main Problems

• Find P(O)computational problem: naive solution is intractablefoward-backward algorithm

• Find the sequence of states that most likely produced observed output O: argmaxQ = P(Q|O)Viterbi algorithm

• for fixed topology find P, B and π that maximize probability to observe OBaum-Welch algorithm

Page 12: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Forward/Backward

• given all parameters (B, Q, π) find P(O)

• naive approach is computationally too intensive

• Use help of forwards α and backwards β

• α(t,i) = P(o1o2 … ot, qt=Si)

• then P(O) = ∑i α(T,i)

Page 13: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Viterbi

• compute arg maxQ P(Q | O)

• two-step algorithm:

• maximize probability first,

• then recover structure Q

Page 14: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Baum-Welch

Page 15: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

HMM for Amino Acid/Gene Sequences

CAEFTPAVH, CKETTPADH, CAETPDDH, CAEFDDH, CDAEFPDDH

d1

i1

m1

d2

i2

m2

d3

i3

m3 m4

i0

m0

Page 16: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Example

• CAEFDDH most likely produced bym0 m1 m2 m3 m4 d5 d6 m7 m8 m9 m10

• CDAEFPDDH most likely produced bym0 m1 i1 m2 m3 m4 d5 m6 m7 m8 m9 m10

• Yields alignment C -AEF - -DDH CDAEFP -DDH

CAEFTPAVH, CKETTPADH, CAETPDDH, CAEFDDH, CDAEFPDDH

Page 17: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

R packages

• HMM

• RHmm

• HiddenMarkov

• msm

• depmix, depmixS4

• flexmix

Page 18: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Bayesian Networks

• HMMs are special case of Bayesian Nets

• A Bayesian Network is a directed acyclic graph, where nodes represent variablesedges describe conditional relationships

Page 19: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Setup

• If there is no edge between two nodes, the nodes are independent

• Edges imply parent/child relationship:

• X1 has children X2, X3, X5

• X5 has parents X1, X2

• P(X1, ..., Xp) = ∏i P(Xi|parents(Xi))

X1

X2

X3

X4X5

Page 20: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

Example

• Given that the grass is wet, what is the probability that it is raining?

Page 21: Hidden Markov Models, Bayesian Networks - Nutritionistahofroe.net/stat430/lectures/28-hidden.pdf · Hidden Markov Models, Bayesian Networks Stat 430. Outline •Definition of HMM

R packages

• deal

• MASTINO