lirong xia hidden markov models tue, march 28, 2014

31
Lirong Xia Hidden Markov Models Tue, March 28, 2014

Upload: emmeline-parker

Post on 16-Dec-2015

220 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Lirong Xia

Hidden Markov Models

Tue, March 28, 2014

Page 2: Lirong Xia Hidden Markov Models Tue, March 28, 2014

• Markov decision process (MDP)

– transition probability only depends on (state,action)

in the previous step

• Reinforcement learning

– unknown probability/rewards

• Markov models

• Hidden Markov models2

The “Markov”s we have learned so far

Page 3: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Markov Models

3

• A Markov model is a chain-structured BN– Conditional probabilities are the same (stationarity)– Value of X at a given time is called the state– As a BN:

– Parameters: called transition probabilities

p(X1) p(X|X-1)

Page 4: Lirong Xia Hidden Markov Models Tue, March 28, 2014

• p(X=sun)=p(X=sun|X-1=sun)p(X=sun)+

p(X=sun|X-1=rain)p(X=rain)

• p(X=rain)=p(X=rain|X-1=sun)p(X=sun)+

p(X=rain|X-1=rain)p(X=rain)4

Computing the stationary distribution

Page 5: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Hidden Markov Models

5

• Hidden Markov models (HMMs)– Underlying Markov chain over state X– Effects (observations) at each time step– As a Bayes’ net:

Page 6: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Example

6

• An HMM is defined by:– Initial distribution: p(X1)

– Transitions: p(X|X-1)

– Emissions: p(E|X)

Rt-1 p(Rt)

t 0.7f 0.3

Rt p(Ut)

t 0.9f 0.2

Page 7: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Filtering / Monitoring

7

• Filtering, or monitoring, is the task of tracking the distribution B(X) (the belief state) over time

• B(Xt) = p(Xt|e1:t)

• We start with B(X) in an initial setting, usually uniform

• As time passes, or we get observations, we update B(X)

Page 8: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Example: Robot Localization

8

Sensor model: never more than 1 mistakeMotion model: may not execute action with small prob.

Page 9: Lirong Xia Hidden Markov Models Tue, March 28, 2014

HMM weather example: a question

s

c r

.1

.2

.6

.3.4

.3

.3

.5

.3

• You have been stuck in the lab for three days (!)• On those days, your labmate was dry, wet, wet,

respectively• What is the probability that it is now raining outside?

• p(X3 = r | E1 = d, E2 = w, E3 = w)

p(w|s) = .1

p(w|c)

= .3 p(w|

r) = .8

Page 10: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Filtering

• Computationally efficient approach: first compute

p(X1 = i, E1 = d) for all states i

• p(Xt, e1:t) = p(et | Xt)Σxt-1 p(xt-1, e1:t-1) p(Xt | xt-1)

s

c r

.1

.2

.6

.3.4

.3

.3

.5

.3

p(w|s) = .1

p(w|c)

= .3 p(w|

r) = .8

Page 11: Lirong Xia Hidden Markov Models Tue, March 28, 2014

• Formal algorithm for filtering

– Elapse of time

• compute p(Xt+1|Xt,e1:t) from p(Xt|e1:t)

– Observe

• compute p(Xt+1|e1:t+1) from p(Xt+1|e1:t)

– Renormalization

• Introduction to sampling

11

Today

Page 12: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Inference Recap: Simple Cases

12

1

1 1 1 1 1

1 1

1 1 1

| , ( , ) = ( ) ( | )

X

p X e p X e p ep x e

p x p e x

1 1|p X e 2p X

1

1

2 1 2

1 2 1

,

= |x

x

p x p x x

p x p x x

Page 13: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Elapse of Time

13

• Assume we have current belief p(Xt-1|evidence to t-1)

B(Xt-1)=p(Xt-1|e1:t-1)

• Then, after one time step passes:

p(Xt|e1:t-1)=Σxt-1p(Xt|xt-1)p(Xt-1|e1:t-1)

• Or, compactly

B’(Xt)=Σxt-1p(Xt|xt-1)B(xt-1)

• With the “B” notation, be careful about – what time step t the belief is about, – what evidence it includes

Page 14: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Observe and renormalization

14

• Assume we have current belief p(Xt| previous

evidence):

B’(Xt)=p(Xt|e1:t-1)

• Then:

p(Xt|e1:t)∝p(et|Xt)p(Xt|e1:t-1)

• Or:

B(Xt) ∝p(et|Xt)B’(Xt)

• Basic idea: beliefs reweighted by likelihood of

evidence

• Need to renormalize B(Xt)

Page 15: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Recap: The Forward Algorithm

15

• We are given evidence at each time and want to know

• We can derive the following updates

1

1

1

1:

1 1:

1

1:

1 1: 1

1 1:1 1

( , )

,

|

, ,

| |

= | | ,

t

t

t

t t X

t t tx

t t t tx

t

t t

t t

t tt t tx

p x e

p x x e

p x x p e x

p e x p x x

p x e

p x e

p x e

We can normalize as we go if we want

to have p(x|e) at each time step, or

just once at the end…

Page 16: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Example HMM

16

Rt-1 p(Rt)

t 0.7f 0.3

Rt-1 p(Ut)

t 0.9f 0.2

Page 17: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Observe and time elapse

17

Observe

Time elapse and renormalize

• Want to know B(Rain2)=p(Rain2|+u1,+u2)

Page 18: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Online Belief Updates

18

• Each time step, we start with p(Xt-1 | previous evidence):

• Elapse of time

B’(Xt)=Σxt-1p(Xt|xt-1)B(xt-1)

• ObserveB(Xt) ∝p(et|Xt)B’(Xt)

• Renormalize B(Xt)

• Problem: space is |X| and time is |X|2 per time step

– what if the state is continuous?

Page 19: Lirong Xia Hidden Markov Models Tue, March 28, 2014

• Real-world robot localization

19

Continuous probability space

Page 20: Lirong Xia Hidden Markov Models Tue, March 28, 2014

20

Sampling

Page 21: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Approximate Inference

21

• Sampling is a hot topic in machine learning, and it’s really simple

• Basic idea:– Draw N samples from a sampling distribution S– Compute an approximate posterior probability– Show this converges to the true probability P

• Why sample?– Learning: get samples from a distribution you don’t know– Inference: getting a sample is faster than computing the

right answer (e.g. with variable elimination)

Page 22: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Prior Sampling

22

+c+s 0.1-s 0.9

-c+s 0.5-s 0.5

|p S C

+c+r 0.8-r 0.2

-c+r 0.2-r 0.8

|p R C

+s

+r+w 0.99

-w 0.01

-r+w 0.90

-w 0.10

-s

+r+w 0.90

-w 0.10

-r+w 0.01

-w 0.99

| ,p W S R

+c 0.5-c 0.5

p C

Samples:

+c, -s, +r, +w

-c, +s, -r, +w

Page 23: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Prior Sampling (w/o evidences)

23

• This process generates samples with probability:

i.e. the BN’s joint probability

• Let the number of samples of an event be

• Then

• I.e., the sampling procedure is consistent

Page 24: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Example

24

• We’ll get a bunch of samples from the BN:+c, -s, +r, +w+c, +s, +r, +w-c, +s, +r, -w+c, -s, +r, +w-c, -s, -r, +w

• If we want to p(W)– We have counts <+w:4, -w:1>– Normalize to get p(W) = <+w:0.8, -w:0.2>– This will get closer to the true distribution with more

samples– Can estimate anything else, too– What about p(C|+w)? p(C|+r,+w)? p(C|-r,-w)?– Fast: can use fewer samples if less time (what’s the

drawback?)

Page 25: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Rejection Sampling

25

• Let’s say we want p(C)– No point keeping all samples around– Just tally counts of C as we go

• Let’s say we want p(C|+s)– Same thing: tally C outcomes, but

ignore (reject) samples which don’t have S=+s

– This is called rejection sampling– It is also consistent for conditional

probabilities (i.e., correct in the limit)

+c, -s, +r, +w+c, +s, +r, +w-c, +s, +r, -w+c, -s, +r, +w-c, -s, -r, +w

Page 26: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Likelihood Weighting

26

• Problem with rejection sampling:– If evidence is unlikely, you reject a lot of samples– You don’t exploit your evidence as you sample – Consider p(B|+a)

• Idea: fix evidence variables and sample the rest

• Problem: sample distribution not consistent!• Solution: weight by probability of evidence given

parents

-b, -a-b, -a-b, -a-b, -a+b, +a

-b, +a-b, +a-b, +a-b, +a+b, +a

Page 27: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Likelihood Weighting

27

+c+s 0.1-s 0.9

-c+s 0.5-s 0.5

|p S C

+c+r 0.8-r 0.2

-c+r 0.2-r 0.8

|p R C

+s

+r+w 0.99

-w 0.01

-r+w 0.90

-w 0.10

-s

+r+w 0.90

-w 0.10

-r+w 0.01

-w 0.99

| ,p W S R

+c 0.5-c 0.5

p C

Samples:

+c, +s, +r, +w

……

Page 28: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Likelihood Weighting

28

• Sampling distribution if z sampled and e fixed evidence

• now, samples have weights

• Together, weighted sampling distribution is consistent

1

, |l

WS i ii

S z e p z Parents Z

1

, |m

i ii

w z e p e Parents E

1 1

, , | |

,

l m

WS i i i ii i

S z e w z e p z Parents Z p e Parents E

p z e

Page 29: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Ghostbusters HMM

29

– p(X1) = uniform

– p(X|X’) = usually move clockwise, but sometimes move in a random direction or stay in place

– p(Rij|X) = same sensor model as before: red means close, green means far away.

1/9 1/9 1/9

1/9 1/9 1/9

1/9 1/9 1/9

p(X1)

1/6 1/6 1/2

0 1/6 0

0 0 0

p(X|X’=<1,2>)

Page 30: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Example: Passage of Time

30

• As time passes, uncertainty “accumulates”

Transition model: ghosts usually go clockwise

T = 1 T = 2 T= 5

Page 31: Lirong Xia Hidden Markov Models Tue, March 28, 2014

Example: Observation

31

• As we get observations, beliefs get reweighted, uncertainty “decreases”

| 'B p ex X B X

Before observation After observation