0% found this document useful (0 votes)
16 views31 pages

4 Ai ML - 2

Uploaded by

sjsjowmm
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
16 views31 pages

4 Ai ML - 2

Uploaded by

sjsjowmm
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 31

Artificial Intelligence and

Machine Learning - 2
What is Learning?
• “Learning is making useful changes in our
minds” – Marvin Minsky
• “Learning is constructing or modifying
representations of what is being
experienced“ – Ryszard Michalski
• “Learning denotes changes in a system that
... enable a system to do the same task more
efficiently the next time” – Herbert Simon

CSCI363 2018 4
Machine Learning
• Machine learning is a subset of AI that learns
to make decisions by fitting mathematical
models to observed data.
Machine Learning?
• Getting computers to program themselves

• Traditional Programming

Data Written Output


Program
• Machine Learning

ML

Data Generated
Output
Program
Types of Learning
• Supervised learning
– Training data includes desired outputs
• Unsupervised learning
– Training data does not include desired outputs
• Reinforcement learning
– Rewards from sequence of actions
Examples of supervised learning
methods
• Supervised
– K Nearest Neighbors (KNN)
– Neural networks
– Bayesian learning
– Decision tree induction
– Support vector machines
– Model ensembles
Examples of unsupervised learning
methods
• Unsupervised
– Clustering such as
a. K-Means,
b. DBScan,
c. Hierarchical clustering
– Dimensionality reduction such as
a. Principal Component Analysis (PCA),
b. Singular Value Decomposition (SVD)
c. Linear Discriminant Analysis (LDA)
Reinforcement learning
• Learn by interacting with a space or an
environment
Reinforcement learning
• A set of states
• A set of actions
• A set of rewards

• Goal: take actions to change the state so that you


receive rewards

• You don’t receive any data – you have to explore


the environment yourself to gather data as you
go
Supervised Learning Notations
• Given examples of a function (X, F(X))
• Predict function F(X) for new examples X
– Discrete F(X): Classification

– Continuous F(X): Regression


Supervised learning Example
• Given x  input, f(x)  output
• Find f for the following problem:
– Disease Diagnosis
• x: properties of patient ( symptoms, lab tests)
• f(x): Disease or recommended therapy
– Face Recognition
• x: bitmap picture of person’s face
• f(x): name of person/ or gender of person
– Automatic Steering
• x: bitmap picture of road surface in front of car
• f(x): Degrees to turn the steering whole
Regression

• Univariate regression problem (one output, real value)


Text classification

• Binary classification problem (two discrete classes)


Music genre classification

• Multiclass classification problem (discrete classes, >2


possible values)
Image classification

• Multiclass classification problem (discrete classes, >2


possible classes)
What is a supervised learning model?

• An equation relating input (age) to output (height)


• Search through family of possible equations to find one that fits training data well
Unsupervised learning
Clustering (unknown categories)

Clustering
(Unsupervised Classification)
Unsupervised learning
• Unsupervised learning
– machine learning algorithm
– used to draw inferences from datasets without labeled
responses.
• Used for exploratory data analysis to find
– hidden patterns or
– grouping in data.
• The clusters are modeled using a measure of
similarity
– which is defined upon metrics such as Euclidean or
probabilistic distance.
Supervised vs. Unsupervised learning
Supervised Unsupervised

need human effort for data no human effort


labeling

learning conditional learning


distribution, i.e., target class distribution of
distribution given features features only
Neural Network
• The architecture of neural
networks is made up of an
input, output, and hidden layer.
• Neural networks themselves, or
artificial neural networks
(ANNs), are a subset of machine
learning designed to mimic the
processing power of a human
brain.
• Neural networks function by Source:
https://commons.wikimedia.org/wiki/File:Artificial_
passing data through the layers neural_network.svg

of an artificial neuron.
Deep learning
• Deep learning is a subset
of machine learning that uses
multilayered neural networks
(ranging from three to several
hundred or thousands), called
deep neural networks.
• Deep learning requires a large
amount of data to train and
are computationally intensive.
However, they can achieve
higher accuracy and perform
better than shallow networks
for complex tasks.
References
• https://www.ibm.com/topics/deep-learning
• https://en.wikipedia.org/wiki/Deep_learning
• https://h2o.ai/wiki/neural-network-
architectures/#:~:text=Standard%20neural%2
0networks&text=Feed%2DForward%20Netwo
rks%20%2D%20A%20multi,way%20to%20the
%20output%20layer.
• https://udlbook.github.io/udlbook/
• CIS 313, AI course in Jouf University
Review Questions
• What is machine learning?
• What are the 3 types of ML? Discuss them.
• List 2 examples of supervised and 2 examples of
unsupervised learning methods.
‫نص‬
• What is reinforcement learning? Write the three sets of
RL?
• What is the x and f(x) for face recognition problem?
• Give two differences between supervised and
unsupervised learning?
• Discuss deep learning.
MCQ
1. What is one of the differences between
machine learning and deep learning?
(a) Data type (b) Use of AI
(c) Complexity of problem solving (d) Algorithm use

2. For face recognition supervised learning problem,


what could be the input?
(a) Image type (b) image pixel
(b) Name of the person (d) gender of the person
Answers
1. C
2. B
True False
1. Machine learning is a part of deep learning.
2. In ML, we fit mathematical models to observed
data.
3. DBScan is an example of supervised learning.
4. In reinforcement learning, the goal is to take actions
to change the state so that you receive rewards.
5. Knn is an example of unsupervised learning.
6. If the output of the supervised learning is
continuous variable, then we call it regression
problem.
Answers
1. False
2. True
3. False
4. True
5. False
6. True

You might also like