Machine Learning

Chapter 11

Machine Learning
What is learning?

Machine Learning
What is learning? That is what learning is. You suddenly understand something you've understood all your life, but in a new way.
(Doris Lessing 2007 Nobel Prize in Literature)

Machine Learning
How to construct programs that automatically improve with experience.

Machine Learning
How to construct programs that automatically improve with experience.

Learning problem:

Task T Performance measure P Training experience E

Machine Learning
Chess game:
Task T: playing chess games Performance measure P: percent of games won against Training experience E: playing practice games againts itself
opponents

Machine Learning
Handwriting recognition:
Task T: recognizing and classifying handwritten words Performance measure P: percent of words correctly Training experience E: handwritten words with given
classifications classified

Designing a Learning System

Choosing the training experience:
Direct or indirect feedback Degree of learner's control Representative distribution of examples

Designing a Learning System

Choosing the target function:
Type of knowledge to be learned Function approximation

Designing a Learning System

Choosing a representation for the target function:
algorithms

Expressive representation for a close function approximation Simple representation for simple training data and learning

10

Designing a Learning System

Choosing a function approximation algorithm
(learning algorithm)

11

Designing a Learning System

Chess game:
Task T: playing chess games Performance measure P: percent of games won against Training experience E: playing practice games againts itself Target function: V: Board R
opponents

12

Designing a Learning System

Chess game:
Target function representation:

V^(b) = w0 + w1x1 + w2x2 + w3x3 + w4x4 + w5x5 + w6x6 x1: the number of black pieces on the board x2: the number of red pieces on the board x4: the number of red kings on the board x3: the number of black kings on the board x5: the number of black pieces threatened by red

x6: the number of red pieces threatened by black

13

Designing a Learning System

Chess game:
Function approximation algorithm:

(<x1 = 3, x2 = 0, x3 = 1, x4 = 0, x5 = 0, x6 = 0>, 100)

x1: the number of black pieces on the board x2: the number of red pieces on the board x4: the number of red kings on the board x3: the number of black kings on the board x5: the number of black pieces threatened by red

x6: the number of red pieces threatened by black

14

Designing a Learning System

What is learning?

15

Designing a Learning System

Learning is an (endless) generalization or induction
process.

16

Designing a Learning System

New problem (initial board)

Experiment Generator

Hypothesis (V^)

Performance System
Solution trace (game history)

Generalizer
Training examples {(b1, V1), (b2, V2), ...}

Critic

17

Issues in Machine Learning

What learning algorithms to be used? How much training data is sufficient? When and how prior knowledge can guide the learning process? What is the best strategy for choosing a next training experience? What is the best way to reduce the learning task to one or more function approximation problems? How can the learner automatically alter its representation to improve its learning ability?

18

Example
Experience
Example 1 2 3 4 Sky Sunny Sunny Rainy Sunny AirTemp Warm Warm Cold Warm Humidity Normal High High High Low Cold High Low Wind Strong Strong Strong Strong Weak Strong Strong Strong Warm Warm Cool Change Same Same ? ? ?
19

Water Warm Warm Warm Cool

Forecast Same Same Change Change

EnjoySport Yes Yes No Yes

Prediction
5 6 7 Rainy Sunny Sunny Warm Warm

Normal

Example
Learning problem:
Task T: classifying days on which my friend enjoys water sport

Performance measure P: percent of days correctly classified Training experience E: days with given attributes and classifications

20

Concept Learning
Inferring a boolean-valued function from training examples of its input (instances) and output (classifications).

21

Concept Learning
Learning problem:
Target function: Hypothesis:
c: X {0, 1}

Target concept: a subset of the set of instances X

Sky AirTemp Humidity Wind Water Forecast {Yes, No}

h: X {0, 1}

Constraints on instance attributes

Characteristics of all instances of the concept to be learned

22

Concept Learning
Satisfaction:
h(x) = 1 iff x satisfies all the constraints of h h(x) = 0 otherwsie

Consistency: Correctness:

h(x) = c(x) for every instance x of the training examples

h(x) = c(x) for every instance x of X

23

Concept Learning
How to represent a hypothesis function?

24

Concept Learning
Hypothesis representation (constraints on instance attributes):
?: any value is acceptable single required value : no value is acceptable
<Sky, AirTemp, Humidity, Wind, Water, Forecast>

25

Concept Learning
General-to-specific ordering of hypotheses:
hj g hk iff xX: hk(x) = 1 hj(x) = 1
Specific
h h h

h1 = <Sunny, ?, ?, Strong, ? , ?> h2 = <Sunny, ?, ?, h3 = <Sunny, ?, ?,

3

? ?

, ? , ?>

General

H
26

Lattice (Partial order)

, Cool, ?>

FIND-S
Example 1 2 3 4 Sky Sunny Sunny Rainy Sunny AirTemp Warm Warm Cold Warm Humidity Normal High High High Wind Strong Strong Strong Strong Water Warm Warm Warm Cool Forecast Same Same Change Change EnjoySport Yes Yes No Yes

h=< ,

h = <Sunny, Warm, Normal, Strong, Warm, Same> h = <Sunny, Warm, h = <Sunny, Warm, ? ? , Strong, Warm, Same> , Strong, ? , ? >

, ,

>

27

FIND-S
Initialize h to the most specific hypothesis in H: For each positive training instance x:
For each attribute constraint ai in h:

Output hypothesis h

Then replace ai by the next more general constraint satisfied by x

If the constraint is not satisfied by x

28

Example 1 2 3 4

Sky Sunny Sunny Rainy Sunny

AirTemp Warm Warm Cold Warm

Humidity Normal High High High

FIND-S
Wind Strong Strong Strong Strong

Water Warm Warm Warm Cool

Forecast Same Same Change Change

EnjoySport Yes Yes No Yes

Prediction
5 6 7

h = <Sunny, Warm,
Rainy Cold

?
High Low

, Strong,
Strong Strong Strong

? ,
Warm Warm Cool

? >
Change Same Same No

Sunny Sunny

Warm Warm

Normal

Yes Yes
29

FIND-S
The output hypothesis is the most specific one that satisfies all positive training examples.

30

FIND-S
The result is consistent with the positive training examples.

31

FIND-S
Is the result is consistent with the negative training examples?

32

FIND-S
Example 1 2 3 4 5 Sky Sunny Sunny Rainy Sunny Sunny AirTemp Warm Warm Cold Warm Warm Humidity Normal High High High Normal Wind Strong Strong Strong Strong Strong Water Warm Warm Warm Cool Cool Forecast Same Same Change Change Change EnjoySport Yes Yes No Yes No

h = <Sunny, Warm,

, Strong,

? ,

? >

33

FIND-S
The result is consistent with the negative training examples if the target concept is contained in H (and the training examples are correct).

34

FIND-S
The result is consistent with the negative training examples if the target concept is contained in H (and the training examples are correct). Sizes of the space:
Size of the instance space: |X| = 3.2.2.2.2.2 = 96 Size of the hypothesis space H = (4.3.3.3.3.3) + 1 = 973 << 296 The target concept (in C) may not be contained in H.
35

Size of the concept space C = 2|X| = 296

FIND-S
Questions:
Has the learner converged to the target concept, as there can be several consistent hypotheses (with both positive and negative training examples)? Why the most specific hypothesis is preferred? What if there are several maximally specific consistent hypotheses? What if the training examples are not correct?

36

List-then-Eliminate Algorithm
Version space: a set of all hypotheses that are consistent with the training examples. Algorithm:
Initial version space = set containing every hypothesis in H

For each training example <x, c(x)>, remove from the version space any hypothesis h for which h(x) c(x) Output the hypotheses in the version space

37

List-then-Eliminate Algorithm
Requires an exhaustive enumeration of all hypotheses in H

38

Compact Representation of Version Space

G (the generic boundary): set of the most generic hypotheses of H consistent with the training data D:
G = {gH | consistent(g, D) H: g >g g consistent(g, D)} g

S (the specific boundary): set of the most specific hypotheses of H consistent with the training data D:
S = {sH | consistent(s, D) H: s >g s consistent(s, D)} s
39

Compact Representation of Version Space

Version space = <G, S> = {hH | gG sS: g g h g s} S

G
40

Example 1 2 3 4

Candidate-Elimination Algorithm
Sunny Sunny Rainy Sunny Sky Warm Warm Cold Warm AirTemp Normal High High High Humidity Strong Strong Strong Strong Wind Warm Warm Warm Cool Water Same Same Forecast Change Change

EnjoySport Yes Yes Yes No

S0 = {<, , , , , >} G0 = {<?, ?, ?, ?, ?, ?>}

S1 = {<Sunny, Warm, Normal, Strong, Warm, Same>} G1 = {<?, ?, ?, ?, ?, ?>} S2 = {<Sunny, Warm, ?, Strong, Warm, Same>} G2 = {<?, ?, ?, ?, ?, ?>}

S3 = {<Sunny, Warm, ?, Strong, Warm, Same>} G3 = {<Sunny, ?, ?, ?, ?, ?>, <?, Warm, ?, ?, ?, ?>, <?, ?, ?, ?, ?, Same>} S4 = {<Sunny, Warm, ?, Strong, ?, ?>} G4 = {<Sunny, ?, ?, ?, ?, ?>, <?, Warm, ?, ?, ?, ?>}
41

Candidate-Elimination Algorithm
S4 = {<Sunny, Warm, ?, Strong, ?, ?>}

<Sunny, ?, ?, Strong, ?, ?>

<Sunny, Warm, ?, ?, ?, ?>

<?, Warm, ?, Strong, ?, ?>

G4 = {<Sunny, ?, ?, ?, ?, ?>, <?, Warm, ?, ?, ?, ?>}

42

Candidate-Elimination Algorithm
Initialize G to the set of maximally general hypotheses in H Initialize S to the set of maximally specific hypotheses in H

43

Candidate-Elimination Algorithm
For each positive example d:
Remove from G any hypothesis inconsistent with d For each s in S that is inconsistent with d:
Remove s from S

Add to S all least generalizations h of s, such that h is consistent with d and some hypothesis in G is more general than h Remove from S any hypothesis that is more general than another hypothesis in S

44

Candidate-Elimination Algorithm
For each negative example d:
Remove from S any hypothesis inconsistent with d For each g in G that is inconsistent with d:
Remove g from G

Add to G all least specializations h of g, such that h is consistent with d and some hypothesis in S is more specific than h Remove from G any hypothesis that is more specific than another hypothesis in G

45

Candidate-Elimination Algorithm
The version space will converge toward the correct target concepts if:
H contains the correct target concept There are no errors in the training examples

A training instance to be requested next should discriminate among the alternative hypotheses in the current version space:

46

Candidate-Elimination Algorithm
Partially learned concept can be used to classify new instances using the majority rule.
S4 = {<Sunny, Warm, ?, Strong, ?, ?>}

<Sunny, ?, ?, Strong, ?, ?>

<Sunny, Warm, ?, ?, ?, ?>

<?, Warm, ?, Strong, ?, ?>

G4 = {<Sunny, ?, ?, ?, ?, ?>, <?, Warm, ?, ?, ?, ?>} 5 Rainy Warm High Strong Cool Same ?
47

Inductive Bias
Size of the instance space: |X| = 3.2.2.2.2.2 = 96 Number of possible concepts = 2|X| = 296 Size of H = (4.3.3.3.3.3) + 1 = 973 << 296

48

Inductive Bias
Size of the instance space: |X| = 3.2.2.2.2.2 = 96 Number of possible concepts = 2|X| = 296 Size of H = (4.3.3.3.3.3) + 1 = 973 << 296 a biased hypothesis space

49

Inductive Bias
An unbiased hypothesis space H that can represent every subset of the instance space X: Propositional logic sentences Positive examples: x1, x2, x3 Negative examples: x4, x5 h(x) (x = x1) (x = x2) (x = x3) x1 x2 x3 h(x) (x x4) (x x5) x4 x5

50

Inductive Bias
x1 x2 x3

x1 x2 x3 x6

x4 x5 Any new instance x is classified positive by half of the version space, and negative by the other half not classifiable

51

Example 1 2 3 4 5 6 7 8 9

Inductive Bias
Day Actor Price Mon Sat Sun Sun Thu Tue Sat Famous Famous Moderate Cheap Moderate Cheap Cheap Cheap Expensive

EasyTicket Yes No No No

Wed

Infamous Infamous Famous Famous Famous Famous Infamous

Expensive Expensive

Yes Yes Yes No ?

10

Sat

Wed

Infamous

Expensive

?
52

Inductive Bias
Example 1 2 Quality Good Bad Price Low High Buy Yes No

3 4

Good Bad

High Low

? ?

53

Inductive Bias
A learner that makes no prior assumptions regarding the identity of the target concept cannot classify any unseen instances.

54

Homework
Exercises
2-1 2.5 (Chapter 2, ML textbook)

55