Machine Intelligence

Preet Kanwal
Associate Professor
Department of Computer Science & Engineering

Acknowledgement : Dr. Rajanikanth K (Former Principal MSRIT, PhD(IISc),

Academic Advisor, PESU)
Prof. K Srinivas (Associate Professor, PESU)
Teaching Assistant (Nishtha V - Sem VII)
Machine Intelligence
Prescribed Textbook
 Machine Intelligence
ISA Evaluation Policy

Assessment Component Conducted For Scaled Down to

ISA-I (CBT) 40 17.5
ISA-II (CBT) 40 17.5
Assignment (4 Quizzes) 80 15
Quiz 1 : Unit 1 (20 Marks)
Quiz 2 : Unit 2 (20 Marks)
Quiz 3 : Unit 3 (20 Marks)
Quiz 4 : Unit 4 & Unit 5 (20 Marks)
Total ISA Marks 50
Machine Intelligence
What are we looking for?

Machine
Intelligence

Machine Artificial
Learning Intelligence
 Machine Intelligence
Motivation

A news item went as follows: ‘Apple buys machine learning firm Perceptio Inc.,
a startup, in an attempt to bring advanced image-classifying artificial
intelligence to smartphones by reducing data overhead which is typically
required of conventional methods’.
https://appleinsider.com/articles/15/10/05/apple-buys-machine-learning-firm-perceptio-smartphone-ai

Sundar Pichai, the CEO of software giant Google, on being asked what is the next
thing at the company, said “I can’t quite tell exactly but advances in AI and
machine learning, we are making a big bet on that. Advances in machine
learning will bring a difference in many many fields.“ while interacting with
students at his alma mater IIT-Kharagpur.
http://imorphosis.com/category/artificial-intelligence
Machine Intelligence
Motivation
 Machine Intelligence
Motivation

● According to the report How AI

Boosts Industry Profits and
Innovations, AI is predicted to
increase economic growth by an
average of 1.7 percent across 16
industries by 2035.
● By 2035, AI technologies could
increase labor productivity by 40
percent or more
● Let us see what our business
leaders have to say about this.
Machine Intelligence
Motivation

https://www.mygreatlearning.com/blog/what-is-arti
ficial-intelligence/
Machine Intelligence
Applications

1. Traffic Alerts (Maps)

● Google Maps is probably THE app we use whenever we go out and
require assistance in directions and traffic.
● But, How does it know which route is the fastest?
● It’s a combination of People currently using the service, Historic Data
of that route collected over time and few tricks acquired from other
companies.
● Everyone using maps provide their location, average speed and the
route in which they are traveling which in turn helps Google collect
massive Data about the traffic, which makes them predict the
upcoming traffic and adjust your route according to it.
Machine Intelligence
Applications

2. Social Media (Facebook)

● Automatic Friend Tagging Suggestions in Facebook or any other social
media platform uses machine learning
● It uses Face detection and Image recognition to automatically find the
face of the person which matches it’s Database and hence suggests us
to tag that person.

3. Transportation and Commuting (Uber)

● UBER uses Machine Learning algorithm layered on top of Historic Trip

Data to make a more accurate ETA prediction.
● How does Uber determine the price of your ride?
● Uber’s biggest uses of Machine Learning comes in the form of surge
pricing, a machine learning model nicknamed as “Geosurge”.
Machine Intelligence
Applications

4. Products Recommendations

● Suppose you check an item on Amazon, but you do not buy it then and there.
● But the next day, you’re watching videos on YouTube and suddenly you see an ad
for the same item. You switch to Facebook, there also you see the same ad.
● So how does this happen?

● This happens because Google tracks your search history, and recommends ads
based on your search history.
Machine Intelligence
Applications

5. Virtual Personal Assistants

Few of the major Applications of Machine Learning here are:

● Speech Recognition
● Speech to Text Conversion
● Natural Language Processing
● Text to Speech Conversion
 Machine Intelligence
Applications

6. Self Driving Cars(Tesla, Waymo)

● Machine Learning plays a very

important role in Self Driving Cars

● Tesla’s current Artificial Intelligence is

driven by hardware manufacturer
NVIDIA, which is based on
Unsupervised Learning Algorithm.

● NVIDIA stated that they didn’t train their model to detect people or any object as such.
● The model works on Deep Learning and it crowdsources data from all of its vehicles and its
drivers. It uses internal and external sensors which are a part of IOT.
Machine Intelligence
Applications

7. Online Video Streaming (Netflix)

● Netflix’s speedy rise has all movie industrialists taken aback – forcing them to
ask, “How on earth could one single website take on Hollywood?”.
● The answer is Machine Learning.

● The Netflix algorithm constantly gathers massive

amounts of data about users’ activities like:
When you pause, rewind, or fast forward or
What day you watch content (TV Shows on
Weekdays and Movies on Weekends) etc.
● They collect this data for each subscriber and
use their Recommender System and a lot of
Machine Learning Applications to provide a
unique experience for each customer.
Machine Intelligence
In our view - What is a Machine?

Digital Computer Processing System

Machine Intelligence
What do we mean by Intelligence?

Two views :
1. Internal View (How humans think)
2. External View (Behavioral Aspect)
 Machine Intelligence
AI Definition/Four Categories Views of AI

● Turing Test
● Understand how humans
● Knowledge, Reasoning, Language,
think and model this process
Learning
● Cognitive Science

Thinking Acting
Humanly Humanly

Thinking
Acting
Rationaly
Rationaly ● Define the right thing as :
Laws of Thought
● Rational implies thinking “Maximizing the goal
or doing the right thing achievement”.
● Use logic to encode the right ● A rational agent achieves the best
thing and process inputs with outcome
framework
 Machine Intelligence
Acting Humanly

For now, programming a computer to pass the Turing test provides plenty to work on.
The computer would need to possess the following capabilities:

● Natural language processing to enable it to communicate successfully in English (or

some other human language);
● Knowledge representation to store information provided before or during the
interrogation;
● Automated reasoning to use the stored information to answer questions and to
draw new conclusions;
● Machine learning to adapt to new circumstances and to detect and extrapolate
patterns.

The issue of acting like a human comes up primarily when AI programs have to interact
with people. These programs must behave according to certain normal conventions of
human interaction in order to make themselves understood. The underlying
representation and reasoning in such a system may or may not be based on a human
model.
Machine Intelligence
To Summarize

Russel & Peter in their AI book delve into four potential goals or
definitions of AI, which differentiates computer systems on the
basis of rationality and thinking vs. acting:

Human approach:
Systems that think like humans
Systems that act like humans

Ideal approach:
Systems that think rationally
Systems that act rationally
 Machine Intelligence
AI Definition

https://www.youtube.com/watch?v=y5swZ2Q_lBw
 Machine Intelligence
AI Definition/Four Categories Views of AI

https://www.youtube.com/watch?v=y5swZ2Q_lBw
 Machine Intelligence
Levels/Stages of Artificial Intelligence

PRESENT FUTURE

Narrow Specialized in General Human Super Smarter than

AI one Task AI Level AI AI Humans
 Machine Intelligence
Artificial Intelligence Vs. Human Intelligence

COMPARISON ARTIFICIAL INTELLIGENCE HUMAN INTELLIGENCE

FACTORS
NATURE Aims to build machines that can Aims to adapt to new environments
mimic human behavior and perform by utilizing a combination of
human-like actions. different cognitive processes,

FUNCTIONING AI-powered machines rely on data Use the brain’s computing power,
and specific instructions fed into the memory, and ability to think,
system.
LEARNING POWER Learn from data and through It is all about learning from various
continuous training, but they can incidents and past experiences. It is
never achieve the thought process about learning from mistakes made
unique to humans via trial and error approach
throughout one’s life.
 Machine Intelligence
AI vs ML

https://www.edureka.co/blog/ai-vs-machine-learning-vs-deep-learning/
 Machine Intelligence
AI vs ML

https://www.mygreatlearning.com/blog/what-is-artificial-intelligence/
Machine Intelligence
Definition of ML
Machine Intelligence
PTE Machine Learning Model Definition

“Learning is any process by which a system improves performance

from experience.”
- Herbert Simon

Definition by Tom Mitchell (1998):

Machine Learning is the study of algorithms that
improve their performance P
• at some task T
• with experience E.

A well-defined learning task is given by <P,T,E>

Machine Intelligence
PTE Machine Learning Model Definition

A checkers learning problem:

● Task T: playing checkers

● Performance measure P:
percent of games won against
opponents

● Training experience E:
playing practice games against
itself
 Machine Intelligence
PTE Machine Learning Model Definition

A handwriting recognition
learning problem:

● Task T: recognizing and

classifying handwritten words
within images
● Performance measure P:
percent of words correctly
classified
● Training experience E: a
database of handwritten words
with given classifications
Machine Intelligence
QUIZ

A computer program is said to learn from experience E with respect to some task T
and some performance measure P if its performance on T, as measured by P,
improves with experience E. Suppose we feed a learning algorithm a lot of historical
weather data, and have it learn to predict weather. What would be a reasonable
choice for P?

The probability of it correctly predicting a future date’s weather.

The weather prediction task.

The process of the algorithm examining a large amount of historical weather data.

None of the above.

Machine Intelligence
QUIZ

A computer program is said to learn from experience E with respect to some task T
and some performance measure P if its performance on T, as measured by P,
improves with experience E. Suppose we feed a learning algorithm a lot of historical
weather data, and have it learn to predict weather. In this setting, what is T?

The weather prediction task

task.

The probability of it correctly predicting a future date’s weather.

The process of the algorithm examining a large amount of historical weather data.

None of these.
Machine Intelligence
Features of Machine Learning
Machine Intelligence
Issues with Machine Learning and AI

● Defining the problem

● Gathering training data
● How much training data is sufficient to learn a concept with high confidence?
● Maintaining machine learning models
● Developing the right machine learning strategy
● What algorithms can approximate functions well and when ?
● When is it useful to use prior knowledge?
● Are some training examples more useful than others?
● What are best tasks for a system to learn?
● What is the best way for a system to represent its knowledge?
● Can the learner automatically alter its representation for improvement ?

At the end of this course you will be able to answer all these questions
 Machine Intelligence
Course Content

Artificial Intelligence
Machine Learning
● Intelligent Agents and its Types
● Problem solving by Searching- ● Performance Metrics
Uninformed Search and Informed ● Concept Learning
Search Methods ● Decision Trees
● KNN, ANN, SVM
● Bayesian Learning, HMM
● K-means clustering,
● Expected Maximization, Gaussian
Deep Learning (Unit 5)
Mixture Models,
● Dimensionality Reduction - PCA,
● CNN SVD
● RNN ● Genetic algorithms
THANK YOU

Preet Kanwal
Department of Computer Science & Engineering
[email protected]