cs 294-5: statistical natural language processingshapiro/ee562/notes/bayesnetsia… · p(toothache,...
TRANSCRIPT
![Page 1: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/1.jpg)
Probabilistic Models
Models describe how (a portion of) the world works
Models are always simplifications May not account for every variable May not account for all interactions between variables “All models are wrong; but some are useful.”
– George E. P. Box
What do we do with probabilistic models? We (or our agents) need to reason about unknown
variables, given evidence Example: explanation (diagnostic reasoning) Example: prediction (causal reasoning) Example: value of information
1
![Page 2: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/2.jpg)
Bayes’ Nets
[These slides were created by Dan Klein and Pieter Abbeel for CS188 Intro to AI at UC Berkeley. All CS188 materials are available at http://ai.berkeley.edu.]
![Page 3: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/3.jpg)
Independence
3
![Page 4: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/4.jpg)
Example: Independence
N fair, independent coin flips:
H 0.5
T 0.5
H 0.5
T 0.5
H 0.5
T 0.5
4
![Page 5: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/5.jpg)
Conditional Independence
P(Toothache, Cavity, Catch)
If I have a cavity, the probability that the probe catches in it doesn't depend on whether I have a toothache: P(+catch | +toothache, +cavity) = P(+catch | +cavity)
The same independence holds if I don’t have a cavity: P(+catch | +toothache, -cavity) = P(+catch| -cavity)
Catch is conditionally independent of Toothache given Cavity: P(Catch | Toothache, Cavity) = P(Catch | Cavity)
Equivalent statements: P(Toothache | Catch , Cavity) = P(Toothache | Cavity) P(Toothache, Catch | Cavity) = P(Toothache | Cavity) P(Catch | Cavity) One can be derived from the other easily
5
![Page 6: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/6.jpg)
Conditional Independence
Unconditional (absolute) independence very rare (why?)
Conditional independence is our most basic and robust form of knowledge about uncertain environments.
DEF: X is conditionally independent of Y given Z
if and only if:
or, equivalently, if and only if
6
![Page 7: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/7.jpg)
Conditional Independence
What about this domain:
Traffic Umbrella Raining
7
U || T | R
u, t, r P(u, t | r) = P(u | r) P(t | r)
R
U T
![Page 8: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/8.jpg)
Conditional Independence
What about this domain:
Fire Smoke Alarm
8
F S A
A || F | S
P(a | s, f) = P(a | s)
![Page 9: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/9.jpg)
Conditional Independence and the Chain Rule
Chain rule:
Trivial decomposition:
With assumption of conditional independence:
Bayes’nets / graphical models help us express conditional independence assumptions9
![Page 10: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/10.jpg)
Ghostbusters Chain Rule
Each sensor depends onlyon where the ghost is
That means, the two sensors are conditionally independent, given the ghost position
T: Top square is redB: Bottom square is redG: Ghost is in the top
Givens:P( +g ) = 0.5P( -g ) = 0.5P( +t | +g ) = 0.8P( +t | -g ) = 0.4P( +b | +g ) = 0.4P( +b | -g ) = 0.8
P(T,B,G) = P(G) P(T|G) P(B|G)
T B G P(T,B,G)
+t +b +g 0.16
+t +b -g 0.16
+t -b +g 0.24
+t -b -g 0.04
-t +b +g 0.04
-t +b -g 0.24
-t -b +g 0.06
-t -b -g 0.0610
e.g. (.5) (.8) (.4) = .16
![Page 11: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/11.jpg)
Bayes’Nets: Big Picture
11
![Page 12: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/12.jpg)
Bayes’ Nets: Big Picture
Two problems with using full joint distribution tables as our probabilistic models: Unless there are only a few variables, the joint is WAY too
big to represent explicitly Hard to learn (estimate) anything empirically about more
than a few variables at a time
Bayes’ nets: a technique for describing complex joint distributions (models) using simple, local distributions (conditional probabilities) More properly called graphical models We describe how variables locally interact Local interactions chain together to give global, indirect
interactions For about 10 min, we’ll be vague about how these
interactions are specified
12
![Page 13: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/13.jpg)
Example Bayes’ Net: Insurance
13
What factor(s) determine
whether there is going to
be an Accident?
What determines
Liability Cost?
![Page 14: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/14.jpg)
Example Bayes’ Net: Car
14
If the car won’t start,
what is the cause?
![Page 15: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/15.jpg)
Graphical Model Notation
Nodes: variables (with domains) Can be assigned (observed) or unassigned
(unobserved)
Arcs: interactions Similar to CSP constraints Indicate “direct influence” between variables Formally: encode conditional independence
(more later)
For now: imagine that arrows mean direct causation (in general, they don’t!)
15
![Page 16: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/16.jpg)
Example: Coin Flips
N independent coin flips
No interactions between variables: absolute independence
X1 X2 Xn
16
![Page 17: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/17.jpg)
Example: Traffic
Variables: R: It rains
T: There is traffic
Model 1: independence
Why is an agent using model 2 better?
R
T
R
T
Model 2: rain causes traffic
17It can plan better.
![Page 18: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/18.jpg)
Let’s build a causal graphical model!
Variables T: Traffic
R: It rains
L: Low pressure
D: Roof drips
B: Ballgame
C: Cavity
Example: Traffic II
R
T
DL
B
C
18
![Page 19: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/19.jpg)
Example: Alarm Network
Variables B: Burglary
A: Alarm goes off
M: Mary calls
J: John calls
E: Earthquake!
19
B E
A
J M
![Page 20: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/20.jpg)
Bayes’ Net Semantics
20
![Page 21: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/21.jpg)
Bayes’ Net Semantics
A set of nodes, one per variable X
A directed, acyclic graph
A conditional distribution for each node
A collection of distributions over X, one for each combination of parents’ values
CPT: conditional probability table
Description of a noisy “causal” process
A1
X
An
A Bayes net = Topology (graph) + Local Conditional Probabilities21
![Page 22: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/22.jpg)
Probabilities in BNs
Bayes’ nets implicitly encode joint distributions
As a product of local conditional distributions
To see what probability a BN gives to a full assignment, multiply all the relevant conditionals together:
Example:
22= P(+cavity)P(-toothache|+cavity)P(+catch|+cavity)
![Page 23: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/23.jpg)
Probabilities in BNs
Why are we guaranteed that setting
results in a proper joint distribution?
Chain rule (valid for all distributions):
Assume conditional independences:
Consequence:
Not every BN can represent every joint distribution
The topology enforces certain conditional independencies 23
![Page 24: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/24.jpg)
Only distributions whose variables are absolutely independent can be represented by a Bayes’ net with no arcs.
Example: Coin Flips
h 0.5
t 0.5
h 0.5
t 0.5
h 0.5
t 0.5
X1 X2 Xn
24
![Page 25: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/25.jpg)
Example: Traffic
R
T
+r 1/4
-r 3/4
+r +t 3/4
-t 1/4
-r +t 1/2
-t 1/2
P(-t | +r) * P(+r)
¼ * ¼
1/16
25
![Page 26: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/26.jpg)
Example: Alarm Network
Burglary Earthqk
Alarm
John calls
Mary calls
B P(B)
+b 0.001
-b 0.999
E P(E)
+e 0.002
-e 0.998
B E A P(A|B,E)
+b +e +a 0.95
+b +e -a 0.05
+b -e +a 0.94
+b -e -a 0.06
-b +e +a 0.29
-b +e -a 0.71
-b -e +a 0.001
-b -e -a 0.999
A J P(J|A)
+a +j 0.9
+a -j 0.1
-a +j 0.05
-a -j 0.95
A M P(M|A)
+a +m 0.7
+a -m 0.3
-a +m 0.01
-a -m 0.99 26
What is
P(b,e,a,j,a) ?
![Page 27: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/27.jpg)
Bayes’ Nets
So far: how a Bayes’ net encodes a joint distribution
Next: how to answer queries about that distribution
27
![Page 28: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/28.jpg)
Example: Traffic
Causal direction
R
T
+r 1/4
-r 3/4
+r +t 3/4
-t 1/4
-r +t 1/2
-t 1/2
+r +t 3/16
+r -t 1/16
-r +t 6/16
-r -t 6/16
28
How do we get the joint
distribution?
P(T,R) = P(R)P(T|R)
+r +t = (3/4) (1/4) = 3/16
-r +t = (1/2) (3/4) = 6/16
![Page 29: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/29.jpg)
Example: Reverse Traffic
Reverse causality?
T
R
+t 9/16
-t 7/16
+t +r 1/3
-r 2/3
-t +r 1/7
-r 6/7
+r +t 3/16
+r -t 1/16
-r +t 6/16
-r -t 6/16
29
Bayes’ Rule
P(R|T) α P(T|R) P(R)
![Page 30: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/30.jpg)
Causality?
When Bayes’ nets reflect the true causal patterns:
Often simpler (nodes have fewer parents) Often easier to think about Often easier to elicit from experts
BNs need not actually be causal
Sometimes no causal net exists over the domain (especially if variables are missing)
E.g. consider the variables Traffic and Drips End up with arrows that reflect correlation, not causation
What do the arrows really mean?
Topology may happen to encode causal structure Topology really encodes conditional independence
30
![Page 31: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/31.jpg)
Bayes’ Net Semantics
A directed, acyclic graph, one node per random variable
A conditional probability table (CPT) for each node
A collection of distributions over X, one for each combination of parents’ values
Bayes’ nets implicitly encode joint distributions
As a product of local conditional distributions
To see what probability a BN gives to a full assignment, multiply all the relevant conditionals together:
31
![Page 32: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/32.jpg)
Size of a Bayes’ Net
How big is a joint distribution over N Boolean variables?
2N
How big is an N-node net if nodes have up to k parents?
O(N * 2k+1)
Both give you the power to calculate
BNs: Huge space savings!
Also easier to elicit local CPTs
Also faster to answer queries (coming)
32
![Page 33: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/33.jpg)
Conditional Independence
X and Y are independent if
X and Y are conditionally independent given Z
(Conditional) independence is a property of a distribution
Example:
33
![Page 34: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/34.jpg)
Bayes Nets: Assumptions
Assumptions we are required to make to define the Bayes net when given the graph:
Beyond above “chain rule Bayes net” conditional independence assumptions
Often additional conditional independences
They can be read off the graph
Important for modeling: understand assumptions made when choosing a Bayes net graph
34
![Page 35: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/35.jpg)
Example
Conditional independence assumptions directly from simplifications in chain rule: X || Z | Y
Y || W | Z
Additional implied conditional independence assumptions?
X Y Z W
35
![Page 36: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/36.jpg)
Independence in a BN
Important question about a BN: Are two nodes independent given certain evidence?
If yes, can prove using algebra (tedious in general)
If no, can prove with a counter example
Example:
Question: are X and Z necessarily independent? Answer: no. Example: low pressure causes rain, which causes traffic.
X can influence Z, Z can influence X (via Y)
Addendum: they could be independent: how?
X Y Z
36
![Page 37: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/37.jpg)
This leades to the concept of D-separation
Study independence properties for triples
Analyze complex cases in terms of member triples
D-separation: a condition / algorithm for answering such queries
37
![Page 38: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/38.jpg)
Causal Chains
This configuration is a “causal chain”
X: Low pressure Y: Rain Z: Traffic
Guaranteed X independent of Z ? No!
One example set of CPTs for which X is not independent of Z is sufficient to show this independence is not guaranteed.
Example:
Low pressure causes rain causes traffic,high pressure causes no rain causes no traffic
(Something else could cause traffic.)
38
X Y Z
![Page 39: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/39.jpg)
Causal Chains
This configuration is a “causal chain” Guaranteed X independent of Z given Y? Proof:
Evidence along the chain “blocks” the influence
Yes!
X: Low pressure Y: Rain Z: Traffic
39
![Page 40: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/40.jpg)
Common Cause
This configuration is a “common cause” Guaranteed X independent of Z ? No!
One example set of CPTs for which X is not independent of Z is sufficient to show this independence is not guaranteed.
Example:
Project due causes both forums busy and lab full
Y: Project due
X: Forums busy
Z: Lab full
40
![Page 41: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/41.jpg)
Common Cause
This configuration is a “common cause” Guaranteed X and Z independent given Y? Proof
Observing the cause blocks influence between effects.
Yes!
Y: Project due
X: Forums busy
Z: Lab full
41
![Page 42: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/42.jpg)
Common Effect
Last configuration: two causes of one effect (v-structures)
Z: Traffic
Are X and Y independent?
Yes: the ballgame and the rain cause traffic, but they are not correlated
Still need to prove they must be (try it!)
Are X and Y independent given Z?
No: seeing traffic puts the rain and the ballgame in competition as explanation.
This is backwards from the other cases
Observing an effect activates influence between
possible causes.
X: Raining Y: Ballgame
42
![Page 43: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/43.jpg)
The General Case
43
![Page 44: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/44.jpg)
The General Case
General question: in a given BN, are two variables independent (given evidence)?
Solution: analyze the graph
Any complex example can be broken
into repetitions of the three canonical cases
44
![Page 45: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/45.jpg)
D-Separation
The Markov blanket of a node is the set of nodes consisting of its parents, its children, and any other parents of its children.
d-separation means directional separation and refers to two nodes in a network.
Let P be a trail (path, but ignore the directions) from node u to v.
P is d-separated by set of nodes Z iff one of the following holds:
1. P contains a chain, um v such that m is in Z
2. P contains a fork, um v such that m is in Z
3. P contains an inverted fork, um v such that m is not in Z and none of its descendants are in Z.
45
![Page 46: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/46.jpg)
Active / Inactive Paths
Question: Are X and Y conditionally independent given evidence variables {Z}?
Yes, if X and Y are “d-separated” by Z
Consider all (undirected) paths from X to Y
No active paths = independence!
Y
X
Active Triples Inactive Triples
Gray nodes are observed.
(the givens)
46
not c-independent
X || Z | Y
X ZY
X || Z | Y
X ZY
X || Z
X Z
Y is not observed
Z
Y
┐(X || Z | Y)
X Z
Y
![Page 47: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/47.jpg)
Active / Inactive Paths
Question: Are X and Y conditionally independent given evidence variables {Z}?
We want NO ACTIVE PATHS.
A path is active if each triple is active:
Causal chain A B C where B is unobserved (either direction)
Common cause A B C where B is unobserved
Common effect (aka v-structure)A B C where B or one of its descendants is observed
Active Triples Inactive Triples
gray nodes are observed.
47
not c-independent
![Page 48: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/48.jpg)
Active / Inactive Paths
Question: Are X and Y conditionally independent given evidence variables {Z}? Yes, if X and Y are “d-separated” by Z Consider all (undirected) paths from X to Y No active paths = independence!
A path is active if each triple is active: Causal chain A B C where B is unobserved (either direction) Common cause A B C where B is unobserved Common effect (aka v-structure)
A B C where B or one of its descendents is observed
All it takes to block a path is a single inactive segment If all paths are blocked by at least one inactive tuple,
then conditionally independent.
Active Triples Inactive Triples
gray nodes are observed.
48
not c-independent
![Page 49: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/49.jpg)
Query:
Check all (undirected!) paths between and
If one or more active, then independence not guaranteed
Otherwise (i.e. if all paths are inactive),
then independence is guaranteed
D-Separation
?
49
![Page 50: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/50.jpg)
Example
Yes R
T
B
T’
No
No
50
![Page 51: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/51.jpg)
Example
R
T
B
D
L
T’
Yes
Yes
Yes
51
No
No
Inactive triple blocks whole path.
L
R
T
B
L->R->T is inactive
R->T<-B is active
One inactive blocks the path
![Page 52: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/52.jpg)
Example
Variables:
R: Raining
T: Traffic
D: Roof drips
S: I’m sad
Questions:
T
S
D
R
Yes
52
No
No
one active path,
and one inactive path
![Page 53: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/53.jpg)
Structure Implications
Given a Bayes net structure, can run d-separation algorithm to build a complete list of conditional independences that are necessarily true of the form
This list determines the set of probability distributions that can be represented
53
![Page 54: CS 294-5: Statistical Natural Language Processingshapiro/EE562/notes/BayesNetsIa… · P(Toothache, Cavity, Catch) If I have a cavity, ... T: Top square is red B: Bottom square is](https://reader033.vdocuments.us/reader033/viewer/2022052803/5f27076582cb1d24e907f241/html5/thumbnails/54.jpg)
Bayes Nets Representation Summary
Bayes nets compactly encode joint distributions
Guaranteed independencies of distributions can be deduced from BN graph structure
D-separation gives precise conditional independence guarantees from graph alone
A Bayes’ net’s joint distribution may have further (conditional) independence that is not detectable until you inspect its specific distribution
54