1. INTRODUCTION TO Machine Learning
Lecture Slides for
INTRODUCTION TO Machine Learning
ETHEM ALPAYDIN
© The MIT Press, 2004

2. CHAPTER 1: Introduction

3. What Is Machine Learning?
Machine learning is programming computers to optimize a performance criterion using example data or past experience.
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

4. Why do we need Machine Learning?
With advance in computer technology, we currently have the ability to store and process large amounts of data, as well as to access it from physically distinct locations over a computer network.
Google=> Special Mail List
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

5. Data Mining
Application of machine learning methods to large databases is called data mining.
Finance banks
credit applications.
Fraud detection.
Stock market.
Manufacturing
Optimization
Control
Troubleshooting
Medicine, telecommunications and World Wide Web.
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

6. Artificial Intelligence
Machine learning is not just a database problem; it is also a part of artificial intelligence.
To be intelligent, a system that is a changing environment should have ability to learn.
If the system can learn and adapt to such changes, the system designer need not foresee and provide solutions for all possible situations.
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

7. A.I.人工智慧
導演: 史蒂芬史匹柏
大衛是一個實驗機器男孩，也是第一個能夠付出真愛的機器人，由一名虛擬電子公司員工和他太太志願領養做為實驗。擁有真愛的機器人無法被人類和機器人接受，大衛不得已帶著他的玩具熊泰迪，踏上尋找自己的定位之旅，終於發現人類和機器之間，存在著無法跨越的鴻溝，但卻也有共通可能，漸漸地大衛被他們視為己出，也獲得寵愛，但是發生一連串意外之後，他卻無法繼續溫暖的家庭生活...
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

8. Pattern Recognition Recognizing faces:
This is a task we do effortlessly; every day we recognize family members and friends by looking at their faces or from their faces or from their photographs, despite difference in pose, lighting, hair style, and so forth.
But we do it unconsciously and are unable to explain how we can do it.
Because we are not able to explain our expertise, we cannot write the computer program.
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

9. Examples of Machine Learning Applications
Learning Associations
Classification (supervised)
Regression (supervised)
Unsupervised Learning
Reinforcement Learning
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

10. Learning Associations
Association rules
P(chips|beer) = 0.7
70 percent of customers who buy beer also buy chips.
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

11. Classification Classification Discrimination Prediction
Saving vs. Income => Low-risk/high-risk.
Prediction
Pattern Recognition
Optical character recognition.
Face recognition
Medical diagnosis
Speech recognition
Knowledge Extraction
Compression
Outlier detection
Fraud.
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

12. Regression Regression Supervised Learning.
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

13. Supervised Learning.
In supervised learning, the aim is to learn a mapping from the input to an output whose correct values are provided by a supervisor.
In unsupervised learning, there is no such supervisor and we only have input of data. The aim is to find the regularities in the input.
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

14. Unsupervised Learning.
In unsupervised learning, there is no such supervisor and we only have input of data.
The aim is to find the regularities in the input.
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

15. Unsupervised Learning
Density Estimation
In Statistics.
Clustering
Image compression
Bioinformatics
DNA=>RNA=>Protein
DNA in our genome is the “blueprint of life”, and is a sequence of bases, namely, A, G, C, and T.
Motif.
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

16. Reinforcement Learning
In such a case, a single action is not important;
What is important is the policy that is the sequence of correct actions to reach the goal.
Game playing
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

17. Why “Learn” ?
Machine learning is programming computers to optimize a performance criterion using example data or past experience.
There is no need to “learn” to calculate payroll
Learning is used when:
Human expertise does not exist (navigating on Mars),
Humans are unable to explain their expertise (speech recognition)
Solution changes in time (routing on a computer network)
Solution needs to be adapted to particular cases (user biometrics)
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

18. What We Talk About When We Talk About“Learning”
Learning general models from a data of particular examples
Data is cheap and abundant (data warehouses, data marts); knowledge is expensive and scarce.
Example in retail: Customer transactions to consumer behavior:
People who bought “Da Vinci Code” also bought “The Five People You Meet in Heaven” (
Build a model that is a good and useful approximation to the data.
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

19. Data Mining
Retail: Market basket analysis, Customer relationship management (CRM)
Finance: Credit scoring, fraud detection
Manufacturing: Optimization, troubleshooting
Medicine: Medical diagnosis
Telecommunications: Quality of service optimization
Bioinformatics: Motifs, alignment
Web mining: Search engines
...
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

20. What is Machine Learning?
Optimize a performance criterion using example data or past experience.
Role of Statistics: Inference from a sample
Role of Computer science: Efficient algorithms to
Solve the optimization problem
Representing and evaluating the model for inference
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

21. Learning
Definition:
A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience.
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

22. Learning Problem Learning: improving with experience at some task
Improve over task T
With respect to performance measure P
Based on experience E
Example: Learn to play checkers:
T: play checkers
P: percentage of games won in a tournament
E: opportunity to play against itself
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

23. Learning Problem Learning: improving with experience at some task
Improve over task T
With respect to performance measure P
Based on experience E
Example: Learn to play checkers:
T: play checkers
P: percentage of games won in a tournament
E: opportunity to play against itself
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

24. Learning to play checkers
T: play checkers
P: percentage of games won
What experience?
What exactly should be learned?
How shall it be represented?
What specific algorithm to learn it?
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

25. Type of Training Experience
Direct or indirect?
Direct: board state -> correct move
Indirect: outcome of a complete game
Credit assignment problem
Teacher or not ?
Teacher selects board states
Learner can select board states
Is training experience representative of performance goal?
Training playing against itself
Performance evaluated playing against world champion
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

26. Choose Target Function
ChooseMove : B  M : board state  move
Maps a legal board state to a legal move
Evaluate : BV : board state  board value
Assigns a numerical score to any given board state, such that better board states obtain a higher score
Select the best move by evaluating all successor states of legal moves and pick the one with the maximal score
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

27. Possible Definition of Target Function
If b is a final board state that is won then V(b) = 100
If b is a final board state that is lost then V(b) = -100
If b is a final board state that is drawn then V(b)=0
If b is not a final board state, then V(b)=V(b’), where b’ is the best final board state that can be achieved starting from b and playing optimally until the end of the game.
Gives correct values but is not operational
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

28. State Space Search V(b)= ? V(b)= maxi V(bi) m3 : bb3 m2 : bb2
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

29. State Space Search V(b1)= ? V(b1)= mini V(bi) m6 : bb6 m5 : bb5
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

30. Final Board States Black wins: V(b)=-100 Red wins: V(b)=100
draw: V(b)=0
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

31. Depth-First Search
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

32. Breadth-First Search
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

33. Number of Board States Tic-Tac-Toe: 4 x 4 checkers: (no queens)
… + 9!/(2! 4! 3!) + … 9 = 6045
4 x 4 checkers: (no queens)
#board states = ?
#board states < 8x7x6x5*22/(2!*2!) = 1680
Regular checkers (8x8 board, 8 pieces each)
#board states < 32!*216/(8! * 8! * 16!) = 5.07*1017
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

34. Choose Representation of Target Function
Table look-up
Collection of rules
Neural networks
Polynomial function of board features
Trade-off in choosing an expressive representation:
Approximation accuracy
Number of training examples to learn the target function
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

35. Determine Representation
Design Choices
Determine Type of
Training Experience
Games against
experts
Games
against self
Table of correct
moves
BoardMove
Determine
Target Function
BoardValue
Determine Representation
of Learned Function
polynomial
Linear function of
six features
Artificial neural
network
Determine
Learning Algorithm
Gradient descent
Linear programming
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

36. Distribution of Applicants
Good customers
Bad customers
Cw=38
Assume we want to minimize classification error:
What is the optimal decision boundary?
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

37. Distribution of Accepted Customers
Cw=43
Good customers
Bad customers
What is the optimal decision boundary?
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

38. Target Function Customer record:
income, owns house, credit history, age, employed, accept
$40000, yes, good, 38, full-time, yes
$25000, no, excellent, 25, part-time, no
$50000, no, poor, 55, unemployed, no
T: Customer data  accept/reject
T: Customer data  probability good customer
T: Customer data  expected utility/profit
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

39. Learning methods Decision rules: Bayesian network: Neural Network:
If income < $ then reject
Bayesian network:
P(good | income, credit history,….)
Neural Network:
Nearest Neighbor:
Take the same decision as for the customer in the data base that is most similar to the applicant
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

40. Learning Problem Examples
Obstacle Avoidance Behavior of a Mobile Robot
Task T: Navigate robot safely through an environment.
Performance measure P : ?
Experience E : ?
Target function : ?
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

41. Performance Measure P:
P: Maximize time until collision with obstacle
P: Maximize distance travelled until collision with obstacle
P: Minimize rotational velocity, maximize translational velocity
P: Minimize error between control action of a human operator and robot controller in the same situation
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

42. Training Experience E:
Direct: Monitor human operator and use her control actions as training data:
E = { <perceptioni, actioni>}
Indirect: Operate robot in the real world or in a simulation. Reward desirable states, penalize undesirable states
V(b) = if v > 0.5 m/s
V(b) = if  < 10 deg/s
V(b) = if bumper state = 1
Question: Internal or external reward ?
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

43. Target Function Choose action: A: perception  action
Sonar readings: s1(t)…sn(t)  <v,>
Evaluate perception/state:
V: s1(t)…sn(t)  V(s1(t)…sn(t))
Problem: states are only partially observable therefore world seems non-deterministic
Markov Decision Process : successor state s(t+1) is a probabilistic function of current state s(t) and action a(t)
Evaluate state/action pairs:
V: s1(t)…sn(t), a(t)  V(s1(t)…sn(t),a(t))
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

44. Learning Methods Neural Networks Reinforcement Learning
Require direct training experience
Reinforcement Learning
Indirect training experience
Evolutionary Algorithms
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

45. Evolutionary Algorithms
10011
01001
mutation
00111
11001
10001
01011
population of genotypes
10111
01001 x f
phenotype space
coding scheme
fitness
selection
11001
10001
01011
recombination
10001
01011
10011
01001 10 01
001
011 10 01
001
011
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

46. Applications Association Supervised Learning Unsupervised Learning
Classification
Regression
Unsupervised Learning
Reinforcement Learning
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

47. Learning Associations
Basket analysis:
P (Y | X ) probability that somebody who buys X also buys Y where X and Y are products/services.
Example: P ( chips | beer ) = 0.7
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

48. Classification Example: Credit scoring
Differentiating between low-risk and high-risk customers from their income and savings
Discriminant: IF income > θ1 AND savings > θ2
THEN low-risk ELSE high-risk
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

49. Classification: Applications
Aka Pattern recognition
Face recognition: Pose, lighting, occlusion (glasses, beard), make-up, hair style
Character recognition: Different handwriting styles.
Speech recognition: Temporal dependency.
Use of a dictionary or the syntax of the language.
Sensor fusion: Combine multiple modalities; eg, visual (lip image) and acoustic for speech
Medical diagnosis: From symptoms to illnesses
...
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

50. Face Recognition Training examples of a person Test images
AT&T Laboratories, Cambridge UK
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

51. Regression Example: Price of a used car x : car attributes y : price
y = g (x | θ )
g ( ) model,
θ parameters
y = wx+w0
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

52. Regression Applications
Navigating a car: Angle of the steering wheel (CMU NavLab)
Kinematics of a robot arm α1 α2
(x,y)
α1= g1(x,y)
α2= g2(x,y)
Response surface design
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

53. Supervised Learning: Uses
Prediction of future cases: Use the rule to predict the output for future inputs
Knowledge extraction: The rule is easy to understand
Compression: The rule is simpler than the data it explains
Outlier detection: Exceptions that are not covered by the rule, e.g., fraud
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

54. Unsupervised Learning
Learning “what normally happens”
No output
Clustering: Grouping similar instances
Example applications
Customer segmentation in CRM
Image compression: Color quantization
Bioinformatics: Learning motifs
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

55. Reinforcement Learning
Learning a policy: A sequence of outputs
No supervised output but delayed reward
Credit assignment problem
Game playing
Robot in a maze
Multiple agents, partial observability, ...
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

56. Scout Robots 16 Sonar sensors Laser range scanner Odometry
Differential drive
Simulator
API in C
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

57. LEGO Mindstorms Touch sensor Light sensor Rotation sensor Video cam
Motors
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

58. Issues in Machine Learning
What algorithms can approximate functions well and when?
How does the number of training examples influence accuracy?
How does the complexity of hypothesis representation impact it?
How does noisy data influence accuracy?
What are the theoretical limits of learnability?
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

59. Machine vs. Robot Learning
Embedded learning
Data distribution not homegeneous
Mostly on-line
Qualitative and sparse feed-back
Time is crucial
Hardware is a priority
Empirical proof
Machine Learning
Learning in vaccum
Statistically well-behaved data
Mostly off-line
Informative feed-back
Computational time not an issue
Hardware does not matter
Convergence proof
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

60. Learning in Robotics behavioral adaptation:
adjust the parameters of individual behaviors according
to some direct feedback signal (e.g. adaptive control)
evolutionary adaptation:
application of artificial evolution to robotic systems
sensor adaptation:
adopt the perceptual system to the environment
(e.g. classification of different contexts, recognition)
learning complex, deliberative behaviors:
unsupervised learning based on sparse feedback from
the environment, credit assignment problem
(e.g. reinforcement learning)
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

61. Resources: Datasets
UCI Repository:
UCI KDD Archive:
Statlib:
Delve:
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

62. Resources: Journals Journal of Machine Learning Research www.jmlr.org
Neural Computation
Neural Networks
IEEE Transactions on Neural Networks
IEEE Transactions on Pattern Analysis and Machine Intelligence
Annals of Statistics
Journal of the American Statistical Association
...
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)

63. Resources: Conferences
International Conference on Machine Learning (ICML)
ICML05:
European Conference on Machine Learning (ECML)
ECML05:
Neural Information Processing Systems (NIPS)
NIPS05:
Uncertainty in Artificial Intelligence (UAI)
UAI05:
Computational Learning Theory (COLT)
COLT05:
International Joint Conference on Artificial Intelligence (IJCAI)
IJCAI05:
International Conference on Neural Networks (Europe)
ICANN05:
...
Lecture Notes for E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1)