COSC 3107: Machine Learning

Lecture 2
Introduction to Machine Learning
COSC-3107 Machine Learning

Shahzad Hussain
Lecturer
Previous Lecture Summary
1. Course Evaluation Detail

2. Introduction to Machine Learning

3. Machine Learning Case Studies

4. Machine Learning Course Journey! COSC-3107 Machine Learning

5. Machine Learning Prerequisites

2 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
Today’s Lecture Outline

6. The State of Art

7. Machine Learning Course Journey!

8. Target Audience
COSC-3107 Machine Learning

9. Machine Learning Prerequisites

10. You will able do amazing things…

3 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
6. The State of the Art
Artificial Intelligence, Machine Learning and Deep Learning

COSC-3107 Machine Learning

Artificial Intelligence, Machine Learning and Deep Learning

• The promise of AI
– Although we may have unrealistic short-term expectations for AI,
the long-term picture is looking bright
• Similarly to Internet (most people didn’t see how the internet was relevant to
them and how it was going to change their lives)
– AI will answer your questions, help educate your kids, and watch
over your health

COSC-3107 Machine Learning

5 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
Conversational Agents
COSC-3107 Machine Learning
 Idea: Conversational Agent
• HAL 9000 is a fictional
artificial intelligence character
and the main antagonist in
Arthur C. Clarke's Space
Odyssey series.
• First appearing in the 1968
film 2001: A Space COSC-3107 Machine Learning
Odyssey, HAL is a sentient
computer that controls the
systems of the Discovery One
spacecraft and interacts
with the ship's astronaut crew

7
Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
Digital Personal Assistants and Chatbots
COSC-3107 Machine Learning
 Digital Personal Assistants and
Chatbots
Google Assistant:
Google Assistant was
launched in 2016 and is
one of the most advanced
chatbots available.
It can be found in a variety
of appliances such as COSC-3107 Machine Learning
telephones,
headphones, speakers,
washers, TVs, and
refrigerators.

9
Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
 Digital Personal Assistants and
Chatbots
Amazon Alexa:
Alexa is a virtual assistant
developed and marketed
by Amazon.
It can interact with users
by voice and by
executing commands COSC-3107 Machine Learning
such as playing music,
creating to-do lists, setting
up alarms, playing
audiobooks, and answering
basic questions.

10
Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
 Digital Personal Assistants and
Chatbots
Apple Siri:
Siri can accept user voice
commands and a natural
language user interface to
answer questions, make
suggestions, and perform actions
by parsing these voice commands
and delegating these requests to a COSC-3107 Machine Learning
set of internet services. The
software can adapt to users'
individual language usage,
their searches, and preferences.
The more it is used the more it
learns and the better it gets.

11
Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
 Digital Personal Assistants and
Chatbots
Microsoft Cortana:
Cortana is another
digital virtual assistant,
designed and created
by Microsoft. Cortana
can set reminders
and alarms,
recognize natural COSC-3107 Machine Learning
voice commands, and
it answers questions
using information.

12
Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
Personal Chauffeur
COSC-3107 Machine Learning
 Google’s Waymo

COSC-3107 Machine Learning

14 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
 Google’s Waymo
• As of 2018, Waymo's autonomous cars have
driven eight million miles on public roads as well
as five billion miles in simulated environments.
• In the next few years, it is all but a certainty that
we will be able to purchase a car capable of full
driving autonomy.
• Tesla, among others, already offers driver
assistance with their Autopilot feature and
possibly will be the first company to offer full self- COSC-3107 Machine Learning

driving capabilities.
• Waymo launched the first commercial driverless
service in 2018 in Arizona, USA with plans to
expand nationally and worldwide.

15 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
Shipping and Warehouse Management
COSC-3107 Machine Learning
Shipping and Warehouse Management
• An Amazon sorting facility is one of the best
examples of the symbiotic relationship that is forming
between humans, computers, and robots.
Computers take customer orders and decide
where to route merchandise, the robots act as
mules carrying the pallets and inventory around the
warehouse.
COSC-3107 Machine Learning
• Even Scott Anderson, Amazon's director of
robotics fulfillment acknowledged in May 2019 that
a fully automated warehouse is at least 10 years
away.

• So, we will continue to see this configuration in

warehouses across the world for a little longer.
17 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
Human Health
COSC-3107 Machine Learning
 Human Health
1. Drug Discovery
2. Insurance pricing
3. Patient Diagnosis
4. Medical Image Interpretation
5. Psychiatric Analysis
6. Smart Health Records
COSC-3107 Machine Learning
7. Disease Detection and Prediction

19 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
Knowledge Search
COSC-3107 Machine Learning
 Knowledge Search
• Google Search, is most often used for
exploring information.

• Early in 2015, as was reported by Bloomberg,

Google began using a deep learning system
called RankBrain to assist in generating
search query responses. COSC-3107 Machine Learning

21 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
Recommendation Systems
COSC-3107 Machine Learning
 Recommendation Systems
• Recommendation systems are another
example of AI technology that has been weaved
into our everyday lives.
• Amazon, YouTube, Netflix, LinkedIn, and
Facebook all rely on recommendation
technology and we don't even realize we are
using it.
COSC-3107 Machine Learning
• Netflix Prize
• Pandora
• Robo-Advisors – betterment

23 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
The Smart Home
COSC-3107 Machine Learning
 The Smart Home
• Home Monitoring:
– Ring Video Doorbell – Amazon
– Google Nest Thermostat
• Vacucmming and mopping
– Roomba
– Brava COSC-3107 Machine Learning
– Packbot – Bomb disposal robot
• Picking up your mess
• Personal Chef
– Moley Robotic Kitchen

25 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
Gaming
COSC-3107 Machine Learning
 StartCraft 2
• StarCraft II is considered one of
the hardest, though it is an ancient
game by video game standards.
• The team at DeepMind introduced
a program dubbed AlphaStar,
defeat a top professional player, in
December 2018, AlphaStar whooped COSC-3107 Machine Learning
a team put together by Grzegorz
"MaNa" Komincz, one of the
world's strongest professional
StarCraft players with a score of 5-
0.

27 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
 IBM Watson –
Question Answering Machine
• IBM and the Watson team
made history in 2011 when
they devised a system that
was able to beat two of the
most successful Jeopardy
champions.
• Watson is a question-
answering system that can COSC-3107 Machine Learning
answer questions posed in
natural language.
• It was initially created by
IBM's DeepQA research
team, led by principal
investigator David Ferrucci.
28 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
 Chess
• The Deep Blue team went back to
the drawing board, made a lot of
enhancements to the software, and
played Kasparov in 1997.
• Deep Blue won the second round
against Kasparov winning the six-
game rematch by a score of 3½– COSC-3107 Machine Learning
2½.
• It then became the first computer
system to beat a current world
champion in a match under standard
chess tournament rules and time
controls.
29 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
 AlphaGo
• There is great example of a
powerful program that can play the
game of Go also developed by
DeepMind called AlphaGo.
• March 2016, it beat Lee Sedol in
a five-game match. This became
the first time a Go program beat a
9-dan professional without
handicaps. Although AlphaGo lost COSC-3107 Machine Learning
to Lee Sedol in the fourth game,
Lee resigned in the final game,
giving a final score
• of 4 games to 1.

30 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
Movie Making
COSC-3107 Machine Learning
 Movie Making
• Deepfakes
– A deepfake is a portmanteau, or blend
of “deep learning” and “fake”.
– It an AI technique to merge video
images.
– Common application is to overlap
someone’s face onto another.
– Used to create fake news or hoaxes.
Example: COSC-3107 Machine Learning
– Zao – developed by Chinese Company
Momo, allows you to overlap someone’s
face over short movie clips like Titanic
and the results are impressive.

32 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
 Movie Making
• Movie Script Generation
– One of the most famous
examples is Sunspring, award
winning.
– Sunspring is an experimental
fiction short film released in
2016. COSC-3107 Machine Learning
– Written by using deep
learning techniques.
– Script was created by LSTM
(Long short-term memory)
model dubbed Benjamin.
33 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
Data Cleaning and Transformation
COSC-3107 Machine Learning
Data Cleaning and Transformation
• Clean and Accurate data is paramount to
producing consistent, reproducible, and accurate AI
models.
• Data scientist spends about 80% of their time
cleaning, preparing and transforming their input
data and 20% for optimizing their models.
• Example: image data sets Image Net and MS COSC-3107 Machine Learning
COCO.
• Amazon Lake formation, August 2019,
automates some steps of creation of a data lake
including collection, cleansing, deduplication,
cataloging, and publication of data.

35 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
7. Machine Learning Course Journey!
COSC-3107 Machine Learning
1-4 Weeks: ML and Applications

ML
Data Intelligence
Method

COSC-3107 Machine Learning

Model & Optimization
Task Evaluation
parameters algorithm

Use P re-specified
(black box)
37 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
4-9 Weeks: Processing, Wrangling and Visualizing Data

• Data Collection
• Data Description
Processing

• Understanding Data
• Filtering Data
Wrangling • Type Casting
• … etc. COSC-3107 Machine Learning

Visualizing • Visualizing with Pandas, Line Charts

• Bar Plots, Histograms, Pie charts
Data • … etc.

38 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
10-12 Weeks: Supervised Learning
• K-Neighbors Models
• Linear Models

• Decision Trees
• Kernelized Support Vector Machines COSC-3107 Machine Learning

• Neural Networks (Deep Learning)

39 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
13-14 Weeks: Unsupervised Learning
• Unsupervised Learning
• Dimensionality Reduction, Feature Extraction and
Manifold Learning
• PCA, NMF and Manifold Learning with t-SNE

• Clustering – K-Means Clustering

COSC-3107 Machine Learning
• Agglomerative Clustering
• DBSCAN

• Neural Networks (Deep Learning)

40 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
15-16 Weeks: Real World Case Studies

• Analyzing Bike Sharing Trends

COSC-3107 Machine Learning

• Galaxy Classification

41 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
16 Weeks: Capstone: Mini-Project and Evaluation
An intelligent application using Deep Learning

Build & Deploy

a recommender using
product images and text COSC-3107 Machine Learning

sentiment

42 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
8. Target Audience
COSC-3107 Machine Learning
Target Audience

Software engineer

Scientist
COSC-3107 Machine Learning

Data enthusiast

44 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
9. Machine Learning Prerequisites
COSC-3107 Machine Learning
Math background

• Basic calculus
- Concept of derivatives
• Basic linear algebra
- Vectors
COSC-3107 Machine Learning
- Matrices
- Matrix Multiply

46 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
Programming Experience
• Basic Python used
- Can pick up along the way if
knowledge of other language

COSC-3107 Machine Learning

47 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
Computing Needs
• Basic D esktop or L aptop
• Access to internet
• Ability to:
- Install and run Python

- Store a few GB of data COSC-3107 Machine Learning

48 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
10. Capstone Project – Semester Project
COSC-3107 Machine Learning
The Capstone Project:
Build and deploy an intelligent
application of deep learning
COSC-3107 Machine Learning
 Intelligent Recommender
using images and text
COSC-3107 Machine Learning

51 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
We will do something more exciting

Text
sentiment Computer
vision
analysis
Capstone
project

COSC-3107 Machine Learning

Recommenders Deep
learning

Deploy
intelligent
web app

52 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology
Today’s Lecture Summary

6. The State of Art

7. Machine Learning Course Journey!

8. Target Audience COSC-3107 Machine Learning

9. Machine Learning Prerequisites

10. Capstone Project – Semester Project

53 Shahzad Hussain, Lecturer, Khawaja Fareed University of Engineering and Information Technology