Machine Learning Basics

1. General Introduction

Compiled For
Ph.D. course Work
APSU, Rewa, MP, India
 Outline

 Artificial Intelligence
 Machine Learning: Modern
Approaches to Artificial Intelligence
 Machine Learning Problems
 Machine Learning Resources
 Our Course

Machine Learning Basics: 1. General Introduction

 Intelligence

 Intelligence
 Ability to solve problems
 Examples of Intelligent Behaviors or
Tasks
 Classification of texts based on content
 Heart disease diagnosis
 Chess playing

Machine Learning Basics: 1. General Introduction

 Example 1: Text Classification (1)

Huge oil platforms dot the Gulf like

beacons -- usually lit up like Christmas
trees at night.
One of them, sitting astride the
Rostam offshore oilfield, was all but Human
blown out of the water by U.S. Judgment
Crude
Warships on Monday.
Ship
The Iranian platform, an unsightly
mass of steel and concrete, was a
three-tier structure rising 200 feet
(60 metres) above the warm waters of
the Gulf until four U.S. Destroyers
pumped some …

Machine Learning Basics: 1. General Introduction

 Example 1: Text Classification (2)

The Federal Reserve is expected to

enter the government securities
market to supply reserves to the
banking system via system repurchase Human
agreements, economists said. Judgment
Most economists said the Fed would Money-fx
execute three-day system
repurchases to meet a substantial
need to add reserves in the current
maintenance period, although some
said a more …

Machine Learning Basics: 1. General Introduction

 Example 2: Disease Diagnosis (1)

Patient 1’s data

Age: 67
Sex: male Doctor
Diagnosis
Chest pain type: asymptomatic
Presence
Resting blood pressure: 160mm Hg
Serum cholestoral: 286mg/dl
Fasting blood sugar: < 120mg/dl
…

Machine Learning Basics: 1. General Introduction

 Example 2: Disease Diagnosis (2)

Patient 2‘s data

Age: 63
Sex: male Doctor
Diagnosis
Chest pain type: typical angina
Absence
Resting blood pressure: 145mm Hg
Serum cholestoral: 233mg/dl
Fasting blood sugar: > 120mg/dl
…

Machine Learning Basics: 1. General Introduction

 Example 3: Chess Playing

 Chess Game
 Two players playing one-by-one under
the restriction of a certain rule
 Characteristics
 To achieve a goal: win the game
 Interactive

Machine Learning Basics: 1. General Introduction

 Artificial Intelligence

 Artificial Intelligence
 Ability of machines in conducting
intelligent tasks
 Intelligent Programs
 Programs conducting specific intelligent
tasks
Intelligent
Processing
Input Output

Machine Learning Basics: 1. General Introduction

 Example 1: Text Classifier (1)

…
…
fiber = 0
Text File: Crude = 1
Preprocessing …
Huge oil Classification …
platforms dot huge = 1
the Gulf like Money-fx = 0
…
beacons -- …
usually lit up … oil = 1
Ship = 1
platforms = 1
…
…

Machine Learning Basics: 1. General Introduction

 Example 1: Text Classifier (2)

…
…
enter = 1
Text File: Crude = 0
Preprocessing expected = 1
The Federal Classification …
Reserve is …
expected to Money-fx = 1
federal = 1
enter the …
government … …
Ship = 0
oil = 0
…
…

Machine Learning Basics: 1. General Introduction

 Example 2: Disease Classifier (1)

Preprocessed data of patient 1

Age = 67
Sex = 1
Classification
Chest pain type = 4
Presence = 1
Resting blood pressure = 160
Serum cholestoral = 286
Fasting blood sugar = 0
…

Machine Learning Basics: 1. General Introduction

 Example 2: Disease Classifier (2)

Preprocessed data of patient 2

Age = 63
Sex = 1
Classification
Chest pain type = 1
Presence = 0
Resting blood pressure = 145
Serum cholestoral = 233
Fasting blood sugar = 1
…

Machine Learning Basics: 1. General Introduction

 Example 3: Chess Program

Searching and
evaluating

Matrix representing Best move -

the current board New matrix

Opponent’s
playing his move

Machine Learning Basics: 1. General Introduction

 AI Approach

 Reasoning with Knowledge

 Knowledge base
 Reasoning
 Traditional Approaches
 Handcrafted knowledge base
 Complex reasoning process
 Disadvantages
 Knowledge acquisition bottleneck

Machine Learning Basics: 1. General Introduction

 Outline

 Artificial Intelligence
 Machine Learning: Modern
Approaches to Artificial Intelligence
 Machine Learning Problems
 Research and Resources
 Our Course

Machine Learning Basics: 1. General Introduction

 Machine Learning

 Machine Learning (Mitchell 1997)

 Learn from past experiences
 Improve the performances of intelligent
programs
 Definitions (Mitchell 1997)
 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 the
tasks improves with the experiences
Machine Learning Basics: 1. General Introduction
 Example 1: Text Classification
Classified text files
Text file 1 trade
Text file 2 ship

… …

Training

New text file Text

class
classifier

Machine Learning Basics: 1. General Introduction

 Example 2: Disease Diagnosis
Database of medical records
Patient 1’s data Absence
Patient 2’s data Presence

… …

Training

New patient’s Disease Presence or

data classifier absence

Machine Learning Basics: 1. General Introduction

 Example 3: Chess Playing
Games played:
Game 1’s move list Win
Game 2’s move list Lose
… …

Training

New matrix Strategy of

representing Searching and Best move
the current
Evaluating
board

Machine Learning Basics: 1. General Introduction

 Examples

 Text Classification
 Task T
 Assigning texts to a set of predefined
categories
 Performance measure P
 Precision and recall of each category
 Training experiences E
 A database of texts with their
corresponding categories
 How about Disease Diagnosis?
 How about Chess Playing?
Machine Learning Basics: 1. General Introduction
 Why Machine Learning Is Possible?

 Mass Storage
 More data available
 Higher Performance of Computer
 Larger memory in handling the data
 Greater computational power for
calculating and even online learning

Machine Learning Basics: 1. General Introduction

 Advantages

 Alleviate Knowledge Acquisition

Bottleneck
 Does not require knowledge engineers
 Scalable in constructing knowledge base
 Adaptive
 Adaptive to the changing conditions
 Easy in migrating to new domains

Machine Learning Basics: 1. General Introduction

 Success of Machine Learning

 Almost All the Learning Algorithms

 Text classification (Dumais et al. 1998)
 Gene or protein classification optionally
with feature engineering (Bhaskar et al.
2006)
 Reinforcement Learning
 Backgammon (Tesauro 1995)
 Learning of Sequence Labeling
 Speech recognition (Lee 1989)
 Part-of-speech tagging (Church 1988)
Machine Learning Basics: 1. General Introduction
 Outline

 Artificial Intelligence
 Machine Learning: Modern
Approaches to Artificial Intelligence
 Machine Learning Problems
 Machine Learning Resources
 Our Course

Machine Learning Basics: 1. General Introduction

 Choosing the Training Experience

 Choosing the Training Experience

 Sometimes straightforward
 Text classification, disease diagnosis
 Sometimes not so straightforward
 Chess playing
 Other Attributes
 How the training experience is controlled
by the learner?
 How the training experience represents
the situations in which the performance
of the program is measured?
Machine Learning Basics: 1. General Introduction
 Choosing the Target Function

 Choosing the Target Function

 What type of knowledge will be learned?
 How it will be used by the program?
 Reducing the Learning Problem
 From the problem of improving
performance P at task T with experience
E
 To the problem of learning some
particular target functions
Machine Learning Basics: 1. General Introduction
 Solving Real World Problems

 What Is the Input?

 Features representing the real world
data
 What Is the Output?
 Predictions or decisions to be made
 What Is the Intelligent Program?
 Types of classifiers, value functions, etc.
 How to Learn from experience?
 Learning algorithms
Machine Learning Basics: 1. General Introduction
 Feature Engineering

 Representation of the Real World Data

 Features: data’s attributes which may be useful
in prediction
 Feature Transformation and Selection
 Select a subset of the features
 Construct new features, e.g.
 Discretization of real value features
 Combinations of existing features
 Post Processing to Fit the Classifier
 Does not change the nature
Machine Learning Basics: 1. General Introduction
 Intelligent Programs

 Value Functions
 Input: features
 Output: value
 Classifiers (Most Commonly Used)
 Input: features
 Output: a single decision
 Sequence Labeling
 Input: sequence of features
 Output: sequence of decisions
Machine Learning Basics: 1. General Introduction
 Examples of Value Functions

 Linear Regression
 Input: feature vectors x  ( x1 , x2 ,, xn )
n
 Output: f (x)  w  x  b   wi xi  b
i 1

 Logistic Regression
 Input: feature vectors x  ( x1 , x2 ,, xn )
1
 Output: f (x) 
1 e
 w x b

Machine Learning Basics: 1. General Introduction

 Examples of Classifiers

 Linear Classifier
 Input: feature vectors x  ( x1 , x2 ,, xn )
n
 Output: y  sgn( w  x  b)  sgn(  wi xi  b)
i 1

 Rule Classifier
 Decision tree
 A tree with nodes representing condition
testing and leaves representing classes
 Decision list
 If condition 1 then class 1 elseif condition 2
then class 2 elseif ….
Machine Learning Basics: 1. General Introduction
 Examples of Learning Algorithms

 Parametric Functions or Classifiers

 Given parameters of the functions or
classifier, e.g.
 Linear functions or classifiers: w, b
 Estimating the parameters, e.g.
 Loss function optimization
 Rule Learning
 Condition construction
 Rules induction using divide-and-conquer

Machine Learning Basics: 1. General Introduction

 Machine Learning Problems

 Methodology of Machine Learning

 General methods for machine learning
 Investigate which method is better under
some certain conditions
 Application of Machine Learning
 Specific application of machine learning
methods
 Investigate which feature, classifier,
method should be used to solve a certain
problem
Machine Learning Basics: 1. General Introduction
 Methodology

 Theoretical
 Mathematical analysis of performances of
learning algorithms (usually with
assumptions)
 Empirical
 Demonstrate the empirical results of
learning algorithms on datasets
(benchmarks or real world applications)

Machine Learning Basics: 1. General Introduction

 Application

 Adaptation of Learning Algorithms

 Directly apply, or tailor learning
algorithms to specific application
 Generalization
 Generalize the problems and methods in
the specific application to more general
cases

Machine Learning Basics: 1. General Introduction

 Outline

 Artificial Intelligence
 Machine Learning: Modern
Approaches to Artificial Intelligence
 Machine Learning Problems
 Machine Learning Resources
 Our Course

Machine Learning Basics: 1. General Introduction

 Introduction Materials

 Text Books
 T. Mitchell (1997). Machine Learning,
McGraw-Hill Publishers.
 N. Nilsson (1996). Introduction to
Machine Learning (drafts).
 Lecture Notes
 T. Mitchell’s Slides
 Introduction to Machine Learning

Machine Learning Basics: 1. General Introduction

 Technical Papers

 Journals, e.g.
 Machine Learning, Kluwer Academic
Publishers.
 Journal of Machine Learning Research,
MIT Press.
 Conferences, e.g.
 International Conference on Machine
Learning (ICML)
 Neural Information Processing Systems
(NIPS)
Machine Learning Basics: 1. General Introduction
 Others

 Data Sets
 UCI Machine Learning Repository
 Reuters data set for text classification
 Related Areas
 Artificial intelligence
 Knowledge discovery and data mining
 Statistics
 Operation research
 …
Machine Learning Basics: 1. General Introduction
 Outline

 Artificial Intelligence
 Machine Learning: Modern
Approaches to Artificial Intelligence
 Machine Learning Problems
 Machine Learning Resources
 Our Course

Machine Learning Basics: 1. General Introduction

 What I will Talk about
 Machine Learning Methods
 Simple methods
 Effective methods (state of the art)
 Method Details
 Ideas
 Assumptions
 Intuitive interpretations

Machine Learning Basics: 1. General Introduction

 What I won’t Talk about
 Machine Learning Methods
 Classical, but complex and not effective
methods (e.g., complex neural networks)
 Methods not widely used
 Method Details
 Theoretical justification

Machine Learning Basics: 1. General Introduction

 What You will Learn
 Machine Learning Basics
 Methods
 Data
 Assumptions
 Ideas
 Others
 Problem solving techniques
 Extensive knowledge of modern
techniques

Machine Learning Basics: 1. General Introduction

 References
 H. Bhaskar, D. Hoyle, and S. Singh (2006). Machine
Learning: a Brief Survey and Recommendations for
Practitioners. Computers in Biology and Medicine, 36(10),
1104-1125.
 K. Church (1988). A Stochastic Parts Program and Noun
Phrase Parser for Unrestricted Texts. In Proc. ANLP-
1988, 136-143.
 S. Dumais, J. Platt, D. Heckerman and M. Sahami
(1998). Inductive Learning Algorithms and
Representations for Text Categorization. In Proc. CIKM-
1998, 148-155.
 K. Lee (1989). Automatic Speech Recognition: The
Development of the Sphinx System, Kluwer Academic
Publishers.
 T. Mitchell (1997). Machine Learning, McGraw-Hill
Publishers.
 G. Tesauro (1995). Temporal Difference Learning and
TD-gammon. Communications of the ACM, 38(3), 58-68.
Machine Learning Basics: 1. General Introduction
The End