introduction to artificial intelligence (ai) · cpsc 502, lecture 9 slide 1 introduction to...
TRANSCRIPT
![Page 1: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/1.jpg)
CPSC 502, Lecture 9 Slide 1
Introduction to
Artificial Intelligence (AI)
Computer Science cpsc502, Lecture 9
Oct, 11, 2011
Slide credit Approx. Inference : S. Thrun, P, Norvig, D. Klein
![Page 2: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/2.jpg)
CPSC 502, Lecture 9 2
Today Oct 11
• Bayesian Networks Approx. Inference
• Temporal Probabilistic Models
Markov Chains
Hidden Markov Models
![Page 3: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/3.jpg)
CPSC 502, Lecture 9 Slide 3
R&Rsys we'll cover in this course
Environment
Problem
Query
Planning
Deterministic Stochastic
Constraint Satisfaction Search
Arc Consistency
Search
Search
Logics
STRIPS
Vars + Constraints
SLS
Value Iteration
Var. Elimination Belief Nets
Decision Nets
Markov Processes
Var. Elimination
Approx. Inference
Temporal. Inference
Static
Sequential
Representation
Reasoning
Technique
![Page 4: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/4.jpg)
Approximate Inference
Basic idea:
Draw N samples from a sampling distribution S
Compute an approximate probability
Show this converges to the true probability P
Why sample?
Inference: getting a sample is faster than computing the right answer (e.g. with variable elimination)
CPSC 502, Lecture 9 4
![Page 5: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/5.jpg)
Prior Sampling
Cloudy
Sprinkler Rain
WetGrass
Cloudy
Sprinkler Rain
WetGrass
5
+c 0.5
-c 0.5
+c
+s 0.1
-s 0.9
-c
+s 0.5 -s 0.5
+c
+r 0.8
-r 0.2
-c
+r 0.2 -r 0.8
+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
Samples:
+c, -s, +r, +w
-c, +s, -r, +w
…
CPSC 502, Lecture 9
![Page 6: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/6.jpg)
Example
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 know 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)?
what’s the drawback? Can use fewer samples ?
Cloudy
Sprinkler Rain
WetGrass
C
S R
W
7
CPSC 502, Lecture 9
![Page 7: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/7.jpg)
Rejection Sampling
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
Cloudy
Sprinkler Rain
WetGrass
C
S R
W
8 CPSC 502, Lecture 9
![Page 8: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/8.jpg)
Likelihood Weighting
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
Burglary Alarm
Burglary Alarm
9
-b, -a
-b, -a
-b, -a
-b, -a
+b, +a
-b +a
-b, +a
-b, +a
-b, +a
+b, +a
CPSC 502, Lecture 9
![Page 9: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/9.jpg)
Likelihood Weighting
10
+c 0.5
-c 0.5
+c
+s 0.1
-s 0.9
-c
+s 0.5 -s 0.5
+c
+r 0.8
-r 0.2
-c
+r 0.2 -r 0.8
+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
Samples:
+c, +s, +r, +w
…
Cloudy
Sprinkler Rain
WetGrass
Cloudy
Sprinkler Rain
WetGrass
CPSC 502, Lecture 9
![Page 10: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/10.jpg)
Likelihood Weighting
Likelihood weighting is good
We have taken evidence into account as
we generate the sample
E.g. here, W’s value will get picked
based on the evidence values of S, R
More of our samples will reflect the state
of the world suggested by the evidence
Likelihood weighting doesn’t solve
all our problems
Evidence influences the choice of
downstream variables, but not upstream
ones (C isn’t more likely to get a value
matching the evidence)
We would like to consider evidence
when we sample every variable 12
Cloudy
Rain
C
S R
W
CPSC 502, Lecture 9
![Page 11: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/11.jpg)
Markov Chain Monte Carlo
Idea: instead of sampling from scratch, create samples
that are each like the last one.
Procedure: resample one variable at a time, conditioned
on all the rest, but keep evidence fixed. E.g., for P(b|+c):
Properties: Now samples are not independent (in fact
they’re nearly identical), but sample averages are still
consistent estimators! And can be computed efficiently
What’s the point: both upstream and downstream
variables condition on evidence. 13
+a +c +b +a +c -b -a +c -b
CPSC 502, Lecture 9
![Page 12: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/12.jpg)
CPSC 502, Lecture 9 14
Today Oct 11
Bayesian Networks Approx. Inference
Temporal Probabilistic Models
Markov Chains
Hidden Markov Models
![Page 13: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/13.jpg)
Modelling static Environments
So far we have used Bnets to perform inference in static
environments
• For instance, the system keeps collecting evidence to
diagnose the cause of a fault in a system (e.g., a car).
• The environment (values of the evidence, the true
cause) does not change as new evidence is gathered
• What does change? The system’s beliefs over possible causes
Slide 15 CPSC 502, Lecture 9
![Page 14: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/14.jpg)
Modeling Evolving Environments: Dynamic Bnets • Often we need to make inferences about evolving
environments.
• Represent the state of the world at each specific point in time via a series of snapshots, or time slices,
Tutoring system tracing student knowledge and morale
Knows-Subtraction t-1
Morale t-1 Morale t
SolveProblem t-1 SolveProblemt
Knows-Subtraction t
Slide 16 CPSC 502, Lecture 9
![Page 15: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/15.jpg)
CPSC 502, Lecture 9 18
Today Oct 11
• Bayesian Networks Approx. Inference
• Temporal Probabilistic Models
Markov Chains
Hidden Markov Models
![Page 16: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/16.jpg)
CPSC 502, Lecture 9 Slide 19
Simplest Possible DBN
• Thus
• Intuitively St conveys all of the information about the
history that can affect the future states.
• “The future is independent of the past given the present.”
• One random variable for each time slice: let’s assume St
represents the state at time t. with domain {s1 …sn }
• Each random variable depends only on the previous one
![Page 17: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/17.jpg)
Simplest Possible DBN (cont’)
• Stationary process assumption: the mechanism that regulates how state variables change overtime is stationary, that is it can be described by a single transition model
• P(St|St-1)
• How many CPTs do we need to specify?
Slide 20 CPSC 502, Lecture 9
![Page 18: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/18.jpg)
Stationary Markov Chain (SMC)
A stationary Markov Chain : for all t >0
• P (St+1| S0,…,St) = P (St+1|St) and
• P (St +1|St) is the same
We only need to specify P (S0) and P (St +1 |St)
• Simple Model, easy to specify
• Often the natural model
• The network can extend indefinitely
• Variations of SMC are at the core of most Natural
Language Processing (NLP) applications!
Slide 21 CPSC 502, Lecture 9
![Page 19: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/19.jpg)
CPSC 502, Lecture 9 Slide 22
Stationary Markov-Chain: Example
Probability of initial state
t
t q p
q
p
0 .3 0
.4 0 .6
0 0 1
a
h
e
0 0 .4
0 0 0
1 0 0
a h e
.3 .4 0
0 0 0
0 0 0
.6 0 0
0 0 1
0 0 0
1tS
tS
Domain of variable Si is {t , q, p, a, h, e}
We only need to specify… t
q
.6
.4
p
a
h
e
0
0
0
0 Stochastic Transition Matrix
P (S0)
P (St+1|St)
![Page 20: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/20.jpg)
CPSC 502, Lecture 9 Slide 23
Markov-Chain: Inference
Probability of a sequence of states S0 … ST
),...,( 0 TSSP
),,( pqtP
Example:
t q
.6
.4 p a h e
0 0 0 0
t
t q p
q
p
0 .3 0
.4 0 .6
0 0 1
a
h
e
0 0 .4
0 0 0
1 0 0
a h e
.3 .4 0
0 0 0
0 0 0
.6 0 0
0 0 1
0 0 0
P (S0)
P (St+1|St)
![Page 21: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/21.jpg)
CPSC 502, Lecture 9 24
Key problems in NLP
Assign a probability to a sentence
• Part-of-speech tagging
• Word-sense disambiguation,
• Probabilistic Parsing
Predict the next word
• Speech recognition
• Hand-writing recognition
• Augmentative communication for the disabled
?),..,( 1 nwwP Impossible to
estimate
?),..,( 1 nwwP“Book me a room near UBC”
Summarization, Machine Translation….....
![Page 22: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/22.jpg)
CPSC 502, Lecture 9 25
Impossible to estimate!
Assuming 105 words in Dictionary and average sentence contains 10 words …….
Google language repository (22 Sept. 2006)
contained “only”: 95,119,665,584 sentences
?),..,( 1 nwwP
Most sentences will not appear or appear only once
![Page 23: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/23.jpg)
CPSC 502, Lecture 9 26
What can we do?
Make a strong simplifying assumption!
Sentences are generated by a Markov Chain
P(The big red dog barks)=
P(The|<S>) *
)|()|(),..,( 12
11 kk
n
kn wwPSwPwwP
![Page 24: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/24.jpg)
CPSC 502, Lecture 9 28
Today Oct 11
• Bayesian Networks Approx. Inference
• Temporal Probabilistic Models
Markov Chains
(Intro) Hidden Markov Models
![Page 25: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/25.jpg)
How can we minimally extend Markov Chains?
• Maintaining the Markov and stationary assumption
A useful situation to model is the one in which:
• the reasoning system does not have access to the states
• but can make observations that give some information about the current state
Slide 29 CPSC 502, Lecture 9
![Page 26: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/26.jpg)
CPSC 502, Lecture 9 Slide 30
Hidden Markov Model
• P (S0) specifies initial conditions
• P (St+1|St) specifies the dynamics
• P (Ot |St) specifies the sensor model
• A Hidden Markov Model (HMM) starts with a Markov
chain, and adds a noisy observation about the state at
each time step:
• |domain(S)| = k
• |domain(O)| = h
![Page 27: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/27.jpg)
CPSC 502, Lecture 9 Slide 31
Example: Localization for “Pushed around” Robot
• Localization (where am I?) is a fundamental problem
in robotics
• Suppose a robot is in a circular corridor with 16
locations
• There are four doors at positions: 2, 4, 7, 11
• The Robot initially doesn’t know where it is
• The Robot is pushed around. After a push it can stay in
the same location, move left or right.
• The Robot has Noisy sensor telling whether it is in front of
a door
![Page 28: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/28.jpg)
CPSC 502, Lecture 9 Slide 32
This scenario can be represented as…
• Example Stochastic Dynamics: when pushed, it stays in the
same location p=0.2, moves left or right with equal probability
P(Loct + 1 | Loc t)
P(Loc1)
![Page 29: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/29.jpg)
CPSC 502, Lecture 9 Slide 33
This scenario can be represented as…
Example of Noisy sensor telling
whether it is in front of a door.
• If it is in front of a door P(O t = T) = .8
• If not in front of a door P(O t = T) = .1
P(O t | Loc t)
![Page 30: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/30.jpg)
Useful inference in HMMs
• Localization: Robot starts at an unknown location and it is pushed around t times. It wants to determine where it is
• In general (Filtering): compute the posterior
distribution over the current state given all
evidence to date
P(Xt | o0:t ) or P(Xt | e0:t )
Slide 34 CPSC 502, Lecture 9
![Page 31: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/31.jpg)
Other HMM Inferences (next time)
• Smoothing (posterior distribution over a past state given all evidence to date)
P(Xk | e0:t ) for 1 ≤ k < t
• Most Likely Sequence (given the evidence seen so far)
argmaxx0:t P(X0:t | e0:t )
![Page 32: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/32.jpg)
CPSC 502, Lecture 9 Slide 36
Also Do exercise 6.E (parts on importance
sampling and particle filtering are optional)
http://www.aispace.org/exercises.shtml
TODO for this Thurs
• Work on Assignment2
• Study the Handout (on approx. inference) Available outside my office after 1pm
![Page 33: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/33.jpg)
CPSC 502, Lecture 9 Slide 37
Example : Robot Localization
• Suppose a robot wants to determine its location based on its
actions and its sensor readings
• Three actions: goRight, goLeft, Stay
• This can be represented by an augmented HMM
![Page 34: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/34.jpg)
CPSC 502, Lecture 9 Slide 38
Robot Localization Sensor and Dynamics Model
• Sample Sensor Model (assume same as for pushed around)
• Sample Stochastic Dynamics: P(Loct + 1 | Actiont , Loc t)
P(Loct + 1 = L | Action t = goRight , Loc t = L) = 0.1
P(Loct + 1 = L+1 | Action t = goRight , Loc t = L) = 0.8
P(Loct + 1 = L + 2 | Action t = goRight , Loc t = L) = 0.074
P(Loct + 1 = L’ | Action t = goRight , Loc t = L) = 0.002 for all other locations L’
• All location arithmetic is modulo 16
• The action goLeft works the same but to the left
![Page 35: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/35.jpg)
CPSC 502, Lecture 9 Slide 39
Dynamics Model More Details
• Sample Stochastic Dynamics: P(Loct + 1 | Action, Loc t)
P(Loct + 1 = L | Action t = goRight , Loc t = L) = 0.1
P(Loct + 1 = L+1 | Action t = goRight , Loc t = L) = 0.8
P(Loct + 1 = L + 2 | Action t = goRight , Loc t = L) = 0.074
P(Loct + 1 = L’ | Action t = goRight , Loc t = L) = 0.002 for all other locations L’
![Page 36: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/36.jpg)
CPSC 502, Lecture 9 Slide 40
Robot Localization additional sensor
• Additional Light Sensor: there is light coming through an
opening at location 10
P (Lt | Loct)
• Info from the two sensors is combined :“Sensor Fusion”
![Page 37: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/37.jpg)
CPSC 502, Lecture 9 Slide 41
The Robot starts at an unknown location and must
determine where it is
The model appears to be too ambiguous
• Sensors are too noisy
• Dynamics are too stochastic to infer anything
http://www.cs.ubc.ca/spider/poole/demos/localization
/localization.html
But inference actually works pretty well.
Let’s check:
You can use standard Bnet inference. However you typically take
advantage of the fact that time moves forward (not in 322)
![Page 38: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/38.jpg)
CPSC 502, Lecture 9 Slide 42
Sample scenario to explore in demo
• Keep making observations without moving. What
happens?
• Then keep moving without making observations.
What happens?
• Assume you are at a certain position alternate
moves and observations
• ….
![Page 39: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/39.jpg)
CPSC 502, Lecture 9 Slide 43
HMMs have many other applications….
Natural Language Processing: e.g., Speech Recognition
• States: phoneme \ word
• Observations: acoustic signal \ phoneme
Bioinformatics: Gene Finding
• States: coding / non-coding region
• Observations: DNA Sequences
For these problems the critical inference is:
find the most likely sequence of states given a
sequence of observations
![Page 40: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/40.jpg)
NEED to explain Filtering
Because it will be used in POMDPs
CPSC 502, Lecture 9 Slide 44
![Page 41: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/41.jpg)
CPSC 502, Lecture 9 Slide 45
Markov Models
Markov Chains
Hidden Markov Model
Markov Decision Processes (MDPs)
Simplest Possible Dynamic Bnet
Add noisy Observations
about the state at time t
Add Actions and Values (Rewards)
![Page 42: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/42.jpg)
CPSC 502, Lecture 9 Slide 46
Answering Query under Uncertainty
Static Belief Network & Variable Elimination
Dynamic Bayesian
Network
Probability Theory
Hidden Markov Models
Email spam filters
Diagnostic
Systems (e.g.,
medicine)
Natural
Language
Processing
Student Tracing in
tutoring Systems Monitoring
(e.g credit cards) BioInformatics Markov Chains
![Page 43: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/43.jpg)
CPSC 502, Lecture 9 Slide 47
Lecture Overview
• Recap
• Temporal Probabilistic Models
• Start Markov Models
Markov Chain
Markov Chains in Natural Language
Processing
![Page 44: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/44.jpg)
CPSC 502, Lecture 9 Slide 48
Sampling a discrete probability
distribution
![Page 45: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/45.jpg)
CPSC 502, Lecture 9 Slide 49
Answering Query under Uncertainty
Static Belief Network & Variable Elimination
Dynamic Bayesian
Network
Probability Theory
Hidden Markov Models
Email spam filters
Diagnostic
Systems (e.g.,
medicine)
Natural
Language
Processing
Student Tracing in
tutoring Systems Monitoring
(e.g credit cards) BioInformatics Markov Chains
![Page 46: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/46.jpg)
CPSC 502, Lecture 9 Slide 50
Answering Queries under Uncertainty
Static Belief Network & Variable Elimination
Dynamic Bayesian
Network
Probability Theory
Hidden Markov Models
Email spam filters
Diagnostic
Systems (e.g.,
medicine)
Natural
Language
Processing
Student Tracing in
tutoring Systems Monitoring
(e.g credit cards) BioInformatics
Markov Chains
Robotics
![Page 47: Introduction to Artificial Intelligence (AI) · CPSC 502, Lecture 9 Slide 1 Introduction to Artificial Intelligence (AI) Computer Science cpsc502, Lecture 9 Oct, 11, 2011 Slide credit](https://reader030.vdocuments.site/reader030/viewer/2022040121/5ecd0f8e2b8a7e25b34f8ad5/html5/thumbnails/47.jpg)
Stationary Markov Chain (SMC)
A stationary Markov Chain : for all t >0
• P (St+1| S0,…,St) = and
• P (St +1|
We only need to specify and
• Simple Model, easy to specify
• Often the natural model
• The network can extend indefinitely
• Variations of SMC are at the core of most Natural
Language Processing (NLP) applications!
Slide 51 CPSC 502, Lecture 9