Introduction to Machine Learning

Ning Xiong

Mälardalen University

1
 Agenda
• What is machine learning?

• Three class of learning problems (supervised,

unsupervised, reinforcement)

• Overview of some learning techniques (linear/logistic

regression, SVM, ANN, k-means clustering)

2
Part 1: What is Machine learning

3
 Inspiration from Human

Learn from experience Learn from data Follow instruction

4
 What is Machine Learning

Machine Learning is to build computer

programs that can improve with experience at
some task

• improve over some Task T

• with respect to a performance measure P
• based on experience E
5
 Applications of ML
• Learning to recognize spoken words
• Learning to drive autonomous vehicles
(Google self-driving car)
• Learning to play world-class games
• Learning to diagnose in medical environments
• Learning for diagnosis and progonosis for
machine maintenance
• Learning for adaptive control of robots or
industrial processes
• ................................
6
Three class of learning problems

7
 Learning Problems
The nature of experience causes different
kinds of learning problems

• Supervised Learning

• Unsupervised Learning

• Reinforcement Learning
8
 Supervised Learning
• A teacher available to give desired response
• Direct experience represented by a set of input-
output examples (xi,yi)
• minimize the error between the actual response of
the learner and the desired response
desired
state x response
Environment Teacher

actual
+
response

S
Learning -
system

error signal 9
 Unsupervised Learning

• No teacher or supervisor available

• No desired outputs in the experiences

• Identify what are normally happening

• Clustering: Grouping similar instances

10
Unsupervised Learning: Clustering
Objects similar in their attributes are clustered in
the same group
Cluster 1

Outlier

x
Cluster 2
11
 Reinforcement Learning
• Aim to obtain optimal action strategy without teacher
• learning through interaction with the environment
• exploration of states and actions
• Indirect experience in form of delayed reward signal
(temporal credit assignment problem)
• goal: maximize accumulated rewards for the long term

Environment
Delayed reward/penalty signal

state
Learning
system
12
action
 Credit Assignment Problem
Assigning credit or blame for the overall outcomes
(delayed reward) to each of the internal decisions
made by the learning machine which contributed to
the overall outcomes

s0 s1 s2 s3 Credit
a1 a2 a3

13
Overview of learning techniques

14
 Linear Regression (Supervised)
Build a linear function to model the relation of dependent
variable Y with its relevant variables

Finding the parameters to best fit the

training samples

å (Y )
n n

å
2
min i - Yˆi = e
ˆ 2
i
i =1 i =1

15
 Logistic Regression (Supervised)
• Build a linear surface to separate training examples into two classes
β0 + β1 X
e
• Convert the linear surface into class probability: π= β0 + β1 X
1+ e
• Finding parameters to minimize the misclassifications, i.e. log loss function

+
+
+
+ -
- -

16
Support Vector Machine (supervised)

Find a linear hyper plane to separate the

training samples, while maximizing the
distances from boundary points to the
plane (maximum margin)

17
 Problem with Linear SVM

=-1
=+1

How to handle when the training examples are not linearly

separable? 18
 Kernel Trick
n General idea: mapping of the original input space to
some higher-dimensional feature space where the
training examples are linearly separable:

Φ: x → φ(x)

19
 Artificial Neural Network
To directly represent complex nonlinear relation, we can build
an artificial neural network consisting of many units.
Output data

Input data
20
 Learning of ANN

Basic idea: gradient descent to learn

the connection weights

• Universal approximator: Implement very complex input-

output functions, approximate any desired function with
arbitrary accuracy.
21
K-Means Clustering (unsupervised)
Initialize: number of clusters K; the initial centroids of the clusters

Iteration: assign data to clusters à recalculate the centroids

22
 Reading Assignment

• Read the first chapter of the ”Machine

Learning” book by Mitchell.

23