Dept.

Computer
Science

Artificial Intelligence
[Intelligent systems]
Rachid Hedjam
CS417
 Why taking CS405?
• As we begin the new millennium
o Science and technology are changing rapidly
o “Old” sciences such as physics are relatively well-understood
o Computers are omnipresent (everywhere)

• Grand Challenges in Science and Technology

o understanding the brain
• reasoning, cognition, creativity
o creating intelligent machines
• is this possible?
• what are the technical and philosophical challenges?
o AI poses the most interesting challenges and questions in
computer science today
R. Hedjam-CS417-Bishop's University
From Professor Padhraic Smyth
Course Learning Outcomes
At the end of this course:
• Knowledge and understanding
Have knowledge and understanding of the basic concepts of Artificial
Intelligence including Search, Uncertainty, Reasoning and Machine
Learning.

• Intellectual skills
You should be able to use this knowledge and understanding of
appropriate principles and guidelines to design solutions to tasks in AI
and to critically evaluate alternatives.

• Practical skills
You should be able to use a Python programming language and to
construct simple AI systems.

R. Hedjam-CS417-Bishop's University
 Today’s Lecture
• What is intelligence?
• What is artificial intelligence?
• A very brief history of AI
• AI in practice
• The rational agent view of AI

R. Hedjam-CS417-Bishop's University
 Introduction

• Can machines think?

• And if so, how?
• And if not, why not?
• And what does this say about human beings?
• And what does this say about the mind?

R. Hedjam-CS417-Bishop's University
 What is artificial intelligence?
• There are no clear consensus on the definition of AI
• Here’s one from John McCarthy, (He coined the phrase AI in 1956)
see http://www. formal. Stanford. EDU/jmc/whatisai/

Q. What is artificial intelligence?

A. It is the science and engineering of making intelligent machines,
especially intelligent computer programs. It is related to the similar
task of using computers to understand human intelligence.
Q. Yes, but what is intelligence?
A. Intelligence is the computational part of the ability to achieve goals in
the world. Varying kinds and degrees of intelligence occur in people,
many animals and some machines.
R. Hedjam-CS417-Bishop's University
Academic Disciplines relevant to AI
• Philosophy • Logic, methods of reasoning, mind as physical system,
foundations of learning, language, rationality.
• Mathematics • Formal representation and proof, algorithms,
computation, (un)decidability, (in)tractability

• Probability/Statisti • modeling uncertainty, learning from data

cs:
• Economics: • utility, decision theory, rational economic agents
• Neuroscience: • neurons as information processing units.
• Psychology/ • how do people behave, perceive, process cognitive
Cognitive Science information, represent knowledge.
• Computer • building fast computers
engineering
• Control theory • design systems that maximize an objective function over
time
• Linguistics • knowledge representation, grammars

R. Hedjam-CS417-Bishop's University
 History of AI
• 1950: Alan Turing publishes Computing Machinery and Intelligence. 'can machines think?' and
introduces the Turing Test to determine if a computer can demonstrate the same intelligence
as a human.
• 1956: John McCarthy coins the term 'artificial intelligence' at the first-ever AI conference at
Dartmouth College. (McCarthy would go on to invent the Lisp language.).
• 1967: Frank Rosenblatt builds the Mark 1 Perceptron, the first computer based on a neural
network that 'learned' through trial and error.
• 1980s: Neural networks which use a backpropagation algorithm to train itself become widely
used in AI applications.
• 1997: IBM's Deep Blue beats then world chess champion Garry Kasparov, in a chess match (and
rematch).
• 2011: IBM Watson beats champions Ken Jennings and Brad Rutter at Jeopardy!
• 2015: Baidu's Minwa supercomputer uses a special kind of deep neural network called a
convolutional neural network to identify and categorize images with a higher rate of accuracy
than the average human.
• 2016: DeepMind's AlphaGo program, powered by a deep neural network, beats Lee Sodol, the
world champion Go player, in a five-game match.
• 2023: A rise in large language models, or LLMs, such as ChatGPT, create an
enormous change in performance of AI and its potential to drive enterprise value.
R. Hedjam-CS417-Bishop's University
https://www.ibm.com/topics/artificial-intelligence
 Four approaches to AI
Human Rational
Creating systems (agents) that Creating systems that think
think like humans rationally
Thinking - Cognitive science - Logic/laws of thought
(Thought)

Creating systems that act like Creating systems that act

humans rationally
- Understand image - Economics/Decision
Acting - Understand natural language
(Behavior) Rational agents

Thinking vs Acting (Acting = Behavior)

Human vs Rational (Rationality R.= Hedjam-CS417-Bishop's
doing the right thing)University
 Intelligent agent
An agent is anything that can be viewed as perceiving its environment
through sensors and acting upon that environment through actuators

Example: Vacuum cleaner

R. Hedjam-CS417-Bishop's University
 Rational agent
• A rational agent is one that does the right thing. It takes the right action for
every perception, leading to maximizing the performance measure, which
makes an agent be the most successful.
• Performance measure: an objective function measuring the performance of
an agent.
– e.g., performance measure of a taxi driver:
• Safe, fast, legal, comfortable trip, maximize profits, etc.
• Rationality: not necessarily “knows everything”.
An agent may not know all the effects of its actions!
An agent does not have to know his entire environment.
An agent is not required to know all the actions from the beginning.
• The automated driver does not need to know what action to take
after the turn (unforeseen!!!). It will adapt by learning from
experience.
• Rationality: not necessarily “perfect”.
– The rationality maximizes the expected (estimated) performance.
– The perfection maximizes the real performance.
• but almost we can't know the real University
R. Hedjam-CS417-Bishop's performance before the action.
 The main topics in AI
• Search (includes Game Playing)
• Knowledge representation and Reasoning
• Learning
• Natural language processing
• Interacting with the Environment (e.g., Vision, Speech recognition,
Robotics)
o to perceive, understand, and act
o e.g., speech recognition and understanding
o e.g., image understanding
o e.g., ability to take actions, have an effect

R. Hedjam-CS417-Bishop's University
 Search
• Search is the fundamental problem of AI.
– Decisions or action plans are abstracted as a search
space, which we then search.
• Search can be:
o Blind (uninformed)
• We move through search space unconcerned with
what comes next, but aware of the answer if we
find it.
o Informed
• We guess what is coming and use that
information (hint or heuristic) to decide where to
move next.
R. Hedjam-CS417-Bishop's University
 Search problems: examples
Short path (google map)

start

Goal

Traveling Salesman problem (delivery)

8-puzzle problem

R. Hedjam-CS417-Bishop's University
 Knowledge Representation &
Reasoning
• Describing (representing) environment and drawing inferences from that
representation.
Sensing

Environment Decision
maker Knowledge
(AI)
Action

We need:
1. A formal language to represent the domain knowledge.
• Logic: makes statements about the world such as describing things: people,
houses, theories, etc.; relations between things and the properties of things.
– All humans are mortal → True statement
2. A method to use the knowledge (Reasoning).
• A mechanism or process of inference that can be used to derive new True
representations (called conclusion) about the world from statements that are
assumed to be True (knowledge base).
• All humans are mortal
• Jean is a human R. Hedjam-CS417-Bishop'sKnowledge
University
base
• Then, Jean is mortal conclusion
 Knowledge Representation &
Reasoning
• Describing (representing) environment and drawing inferences from that
representation.
Sensing

Environment Decision
maker Knowledge
(AI)
Action

We need:
1. A formal language to represent the domain knowledge.
• Logic: makes statements about the world such as describing things: people,
houses, theories, etc.; relations between things and the properties of things.
– All humans are mortal → True statement
2. A method to use the knowledge (Reasoning).
• A mechanism or process of inference that can be used to derive new True
representations (called conclusion) about the world from statements that are
assumed to be True (knowledge base).
• All humans are mortal Limited to
Knowledge
• Jean is a human R. Hedjam-CS417-Bishop's University base T/F facts
• Then, Jean is mortal conclusion
 Learning
• An agent is learning if it improves its performance on future
tasks after making observations about the world.
• Agent learns from experience (data).

(Illustration: Noémie Frezel-​Jacob / D-​INFK)

• Many forms of learning.
– Supervised learning
– Unsupervised learning
– Semi-supervised learning
– Reinforcement learning
– Collaborative learning Supervised learning
– Etc.
• Next topic: Supervised learning. From a collection of input–
output pairs, learn a function (model) that predicts the
output for new inputs.
R. Hedjam-CS417-Bishop's University
 Interacting with the
Environment
• Enabling Intelligent behavior requires interaction with the
environment.
• Intelligent systems can:
o acquire input from sensors: vision, sound, …→ Computer vision
o understand language, recognize speech,
generate text, speech, … → NLP
o modify the environment → Robotics

R. Hedjam-CS417-Bishop's University
 AI Applications
• Autonomous rovers and Satellite stations
 AI Applications
• Medicine: Tumor detection

Image credit: Brain Tumor Segmentation with Deep Neural Networks

R. Hedjam-CS417-Bishop's University
 AI Applications
• Transportation: Autonomous vehicle control:

https://www.eescorporation.com/

R. Hedjam-CS417-Bishop's University
 AI Applications
• Security: Pedestrian detection:

https://data-flair.training/blogs/pedestrian-detection-python-opencv/

R. Hedjam-CS417-Bishop's University
 AI Applications
• Games

https://www.engati.com/blog/ai-in-gaming

https://www.freecodecamp.org/news/simple-chess-ai-step-by-step-1d55a9266977/

R. Hedjam-CS417-Bishop's University
 AI Applications
• Chatbot

https://siamcomputing.com/digital-transformation/chatbot/
R. Hedjam-CS417-Bishop's University
 References
• Artificial Intelligence A Modern Approach Third Edition, Stuart
J. Russell and Peter Norvig (Chapt. 1 & 2)
• Online resources

R. Hedjam-CS417-Bishop's University