Professional Documents
Culture Documents
Chapter 4 ML PDF
Chapter 4 ML PDF
Machine Learning
Dr. Mohamad Hafis Izran Bin Ishak
Department of Control & Mechatronics (CMED)
Faculty of Electrical Engineering
Universiti Teknologi Malaysia.
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global 3
Machine Learning is…
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Machine Learning is…
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Machine Learning is…
past future
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Show Time
• Machine Learning (ML)
• Introduction
• What is ML?
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Algorithms
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Types of Learning Algorithm
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Algorithms
Semi-supervised learning
10
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Show Time
• Machine Learning (ML)
• Supervised Learning
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Learning techniques
- Supervised Learning
• Supervised learning categories and techniques:
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Learning techniques
- Supervised Learning
• Linear classifier
• Techniques:
– Perceptron
– Logistic regression
– Support vector machine (SVM)
– Ada-line
– Multi-layer perceptron (MLP)
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Data
examples
Data
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Supervised learning
examples
label
label1
label2
labeled examples
label3
label4
label
label1
model/
label2 predictor
label3
label4
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Supervised learning
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Classification Example
Differentiate
between low-risk
and high-risk
customers from
their income and
savings
Face recognition
Character recognition
Spam detection
...
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Face Recognition
Test images
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global 20
Regression Example
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global 21
Regression Applications
Epidemiology
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Show Time
• Machine Learning (ML)
• Unsupervised Learning
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Learning techniques
- Unsupervised Learning
• Unsupervised learning categories and techniques
– Clustering
• K-means clustering
• Spectral clustering
– Density Estimation
• Gaussian mixture model (GMM)
• Graphical models
– Dimensionality reduction
• Principal component analysis (PCA)
• Factor analysis
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Unsupervised learning
image compression
…
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Learning techniques
- Reinforcement Learning
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Reinforcement learning
example
Backgammon
… WIN!
… LOSE!
Type of model:
• generative vs. discriminative
• parametric vs. non-parametric
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global
Homework (2 marks each)
– What is it?
– Advantages & Disadvantages
– Example Applications
Dr. Hafis
Herman SKEM4173
MKEM 1713 innovative • entrepreneurial • global