Introduction to Machine Learning Introduktion til maskinlring - - PowerPoint PPT Presentation

▶
introduction to machine learning
SMART_READER_LITE
LIVE PREVIEW

Introduction to Machine Learning Introduktion til maskinlring - - PowerPoint PPT Presentation

Jul 02, 2023 251 likes •442 views

DM825 (5 ECTS - 4th Quarter) Introduction to Machine Learning Introduktion til maskinlring DM825 Machine Learning - L0 Marco Chiarandini adjunkt, IMADA www.imada.sdu.dk/~marco/ 1 Machine Learning A computer program is said to learn from

slide-1
SLIDE 1

DM825 Machine Learning - L0

Marco Chiarandini

adjunkt, IMADA www.imada.sdu.dk/~marco/

DM825 (5 ECTS - 4th Quarter) Introduction to Machine Learning

Introduktion til maskinlœring

1

slide-2
SLIDE 2

DM825 Machine Learning - L0

Machine Learning

Tom M. Mitchell (1997) Machine Learning p.2

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 E.

2

slide-3
SLIDE 3

DM825 Machine Learning - L0

Machine Learning

Core objective of a learner: generalize from its experience. Training examples from experience come from unknown probability distribution. The learner has to extract something to produce a useful answer in new cases.

Tom M. Mitchell (1997) Machine Learning p.2

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 E.

2

slide-4
SLIDE 4

DM825 Machine Learning - L0

Contents

  • Classification and Regression via Linear Models
  • Neural Networks
  • Graphical Models

Bayesian Networks Hidden Markov Models

  • Mixture Models and Expectation Maximization
  • Support Vector Machines
  • Assessment and Selection
  • Unsupervised Learning

(Association rules, cluster analysis, principal components)

3

slide-5
SLIDE 5

DM825 Machine Learning - L0

Perceptron algorithm

4

slide-6
SLIDE 6

DM825 Machine Learning - L0

Multilayered Neural Networks

5

slide-7
SLIDE 7

DM825 Machine Learning - L0

Applications

6

slide-8
SLIDE 8

DM825 Machine Learning - L0

Applications

Handwritten digit recognition

Humans are at 0.2% – 2.5 % error 400–300–10 unit MLP = 1.6% error LeNet: 768–192–30–10 unit MLP = 0.9% error

7

slide-9
SLIDE 9

DM825 Machine Learning - L0

Graphical Models

Allow to represent our prior knoweldge and to use a general suite of algorithms to make inference and to improve our models for a specific application domain Complex systems involve uncertainty => Probability framework interralated aspects of the system are modelled as random variables

8

slide-10
SLIDE 10

DM825 Machine Learning - L0

Example: Medical diagnosis

  • two deases: Fly and Hayfever
  • they are not mutually exclusive
  • Season might be correlated with them
  • symptoms such as Congestion and Muscle Pain

4 random variables: Flu = {true,false}; Hayfever = {true, false} Season = {fall, winter, spring, summer} Congestion = {true, false} MusclePain = {true, false}

P(Flu=true | Season=fall, Congestion=true, MusclePain=false) If the number of variables grows the problem becomes intractable 2x2x4x2x2=64 possible prob. values for joint distribution

9

slide-11
SLIDE 11

DM825 Machine Learning - L0

Example: continued

Graphical models use graph-based representation to encode independencies Season Flu MusclePain Hayfever Congestion P(S,F,H,C,M)=P(S)P(F|S)P(H|S)P(C|F,H)P(M|F)

F and H independent given Season C and S independent given F and H M and H,C independent given F M and C independent gien F

We thus only need to define 3+ 4 +4 +4 +2 =17 parameteers

10

slide-12
SLIDE 12

DM825 Machine Learning - L0

Bayesian Learning

What can we do from here?

  • Inference: Complexity issues O(2^n)
  • Learning (parameters and structure)

11

slide-13
SLIDE 13

DM825 Machine Learning - L0

Bayesian Learning

What can we do from here?

  • Inference: Complexity issues O(2^n)
  • Learning (parameters and structure)

Thumbtack Experiment

11

slide-14
SLIDE 14

DM825 Machine Learning - L0

Bayesian Learning

What can we do from here?

  • Inference: Complexity issues O(2^n)
  • Learning (parameters and structure)

Thumbtack Experiment

Flip the thumbtack in the air and observe the number of times it lands with head and tail We wish to learn how much the probability deviates from 0.5

11

slide-15
SLIDE 15

DM825 Machine Learning - L0

Bayesian Learning

What can we do from here?

  • Inference: Complexity issues O(2^n)
  • Learning (parameters and structure)

Thumbtack Experiment

Flip the thumbtack in the air and observe the number of times it lands with head and tail We wish to learn how much the probability deviates from 0.5

11

slide-16
SLIDE 16

DM825 Machine Learning - L0

Bayesian Learning

What can we do from here?

  • Inference: Complexity issues O(2^n)
  • Learning (parameters and structure)

Thumbtack Experiment

Flip the thumbtack in the air and observe the number of times it lands with head and tail We wish to learn how much the probability deviates from 0.5

Suppose we observe 3 heads in 10 tosses.

  • With no prior knowledge we would set p=3/10=0.33
  • With a prior of 10 heads over 20 tosses we would set p=(3+10)/

(10+20)=13/30=0.43

  • However if we obtain more data the effect diminshes:

(300+1)/1000+2=0.3 and (300+10)/(1000+20)=0.3

11

slide-17
SLIDE 17

DM825 Machine Learning - L0

Prerequisites

✓ MM501 Calculus I ✓ MM505 Linear Algebra ✓ Basics of Probability Calculus

Course Organization

Final Assessment (5 ECTS)

  • Mandatory assignments, pass/fail, internal

evaluation by the teacher. Include programming work in R

  • 3 hours written exam, Danish 7 mark scale
  • External examiner

12

slide-18
SLIDE 18

DM825 Machine Learning - L0

Course Material

  • Text book
  • C.M. Bishop. Pattern recognition and

Machine Learning Springer, 2006

  • Slides
  • Source code and data sets
  • www.imada.sdu.dk/~marco/DM825

13