Lesson 1

Introduction to Artificial intelligence –

Definition of AI, foundations of AI, history of AI, state of the art of
AI, risks and benefits of AI
 Outline

• What Is AI?
• The Foundations of Artificial Intelligence
• The History of Artificial Intelligence
• The State of the Art
• Risks and Benefits of AI
 What is AI?

• Artificial intelligence (AI), in its broadest sense, is intelligence

exhibited by machines, particularly computer systems. It is a
field of research in computer science that develops and
studies methods and software that enable machines to
perceive their environment and use learning and intelligence
to take actions that maximize their chances of achieving
defined goals.
- Wikipedia
What is AI?

- Artificial Intelligence: A Modern Approach Fourth Edition

What is AI?

- Artificial Intelligence: A Modern Approach Third Edition

 Thinking humanly: Cognitive
Modeling Approach

• The 1960s “cognitive revolution”

• Definition: This approach aims to build AI systems that think
like humans. It involves modeling the cognitive processes of
the human mind.
• Key techniques: Cognitive science, neuroscience, psychology,
and creating computational models that mimic human
thought processes.
• Applications: Creating an AI that simulates how humans solve
puzzles by replicating human-like thought patterns.
 Cognitive Science, Cognitive
Neuroscience, and Psychology
• Cognitive Science
– The study of the human mind and brain, focusing on how the mind represents and
manipulates knowledge and how mental representations and processes are
realized in the brain.
– Cognitive science has emerged at the interface of several disciplines.
– As a consequence of this diverse ancestry, cognitive science incorporates a variety
of perspectives and methodologies
• Cognitive Neuroscience
– a subfield of neuroscience that studies the biological processes that underlie
human cognition, especially in regards to the relation between brain structures,
activity, and cognitive functions.
– The purpose is to determine how the brain functions and achieves performance
• Psychology
– The scientific study of behavior and mental processes.
– It aims to understand how people think, feel, and act, both individually and in
groups.
 Acting humanly: Turing Test
Approach

• Definition: This approach aims to build AI systems that can behave

indistinguishably from humans. The goal is for AI to pass the Turing Test,
where a human evaluator cannot distinguish between responses from a
human and an AI.
• Key techniques: Natural language processing, knowledge representation,
machine learning, robotics, computer vision, and other techniques to
replicate human behavior and interaction.
• Applications: Chatbots and virtual agents designed to engage in human-
like conversations, computer vision, robotics, and so on.
 Turing Test

• The Turing Test is a measure of a machine's ability to exhibit intelligent

behavior equivalent to, or indistinguishable from, that of a human
• Alan Turing in his 1950 paper "Computing Machinery and Intelligence."
The test aims to answer the question, "Can machines think? “
– Can machines think?” → “Can machines behave intelligently?”
– Predicted that by 2000, a machine might have a 30% chance of fooling a lay
person
 Thinking rationally: Laws of
Thought Approach

• Definition: It is focused on creating systems that reason logically and make

decisions based on formal rules and principles.
• Key techniques: This approach leverages symbolic AI, logic programming,
and formal methods to achieve reasoning capabilities that align with human
logic.
• Applications: An AI system that uses logical inference to solve
mathematical problems or to derive conclusions from a set of premises
– Logical reasoning
– Expert systems
– Planning and decision making
– Formal verification
 Acting rationally: Agent Approach

• Definition: The "acting rationally" approach to AI, also known as the

rational agent approach, focuses on designing agents that act to achieve the
best possible outcomes based on their goals and the information available to
them.
• Key techniques: Core principles are rationality, performance measure,
environments. This approach leverages principles from search, decision
theory, game theory, reinforcement learning, and optimization.
• Applications: self-driving cars, drones, robotics, healthcare, finance,
gaming, energy management and so on.
 The Foundations of AI
• Philosophy
• Mathematics
• Economics
• Neuroscience
• Cognitive science
• Cognitive neuroscience
• Psychology
• Computer engineering
• Computer science
• Control theory
• Cybernetics
• Linguistics
 History of AI
• 1940s-1950s: Early days: neuron and primitive neural
– 1943: McCulloch & Pitts: Perceptron–boolean circuit model of brain
• 1940s-1950s: Early days: neural and computer science meet
– 1950: Turing's “Computing Machinery and Intelligence”
– 1956: The term "artificial intelligence" was coined.
– 1959: LISP
• 1960s: Early Enthusiasm and Research
– 1961: Unimate
– 1965: DENDRAL
– 1965: ELIZA
– 1966: Shakey
– 1969: Checkers-playing program
• 1970s: Early Development, Specialized Systems, and the first AI Winter
– 1970: Marvin Minsky and Seymour Papert published "Perceptrons,"
– 1972: PROLOG
– 1972: MYCIN
– 1979: XCON
– Late 1970s: The first AI Winter
 History of AI
• 1980s: Expert Systems and the second AI Winter,
– 1980s: Widespread adoption of expert systems
– 1986: Backpropagation
– 1987 - 1993: The second AI Winter
• 1990s: Revival and Emergence of Machine Learning and Early Successes
– 1990: Hidden Markov Models
– 1997: IBM’s Deep Blue
– Late 1990s: Support Vector Machines
• 2000s: Data-Driven AI, Early Deep Learning and the Genesis of Generative AI
– 2000: Bayesian Networks
– 2005: DARPA Grand Challenge
– 2006: Deep Learning
– 2009: Waymo
– 2000s: Big Data
 History of AI
• 2010s: Rise of AI and Breakthroughs
– 2011: IBM Watson
– 2012: AlexNet
– 2014: GANs
– 2016: AlphaGo
– 2017: Transformer
– 2010s : Multimodal systems
• 2020s: AI Boom - Pervasive AI and New Horizons
– 2020: GPT-3
– 2021: AlphaFold2
– 2022: DALL-E2
– 2022: ChatGPT-3.5
– 2023: ChatGPT-4
– 2024: Continued integration of AI in various industries
Reflection of the development
history of AI
 The AI Index Report 2024
• A comprehensive annual report
that provides a detailed overview
of the current state and trends in
artificial intelligence research,
development, and deployment
• https://aiindex.stanford.edu/
 State of Art in AI
• Generative AI
• Reinforcement Learning
• Robotics and Autonomous Systems
• AI in Healthcare
• Natural Language Processing
• AI Ethics and Fairness
• AI Hardware
• Ethical and Policy Frameworks
• AI in Creative Arts
• Edge AI
 Artificial General Intelligence (AGI)

• A form of artificial intelligence that is capable of understanding, learning,

and applying intelligence across a broad range of tasks at a level
comparable to human intelligence.
• AGI encompasses and integrates the four categories of AI—thinking
humanly, acting humanly, thinking rationally, and acting rationally.
• Key Characteristics of AGI
– Generalization
– Human-Level Cognitive Abilities
– Autonomy
– Reasoning and Problem-Solving
– Human-Like Interaction
 Benefits and Risks of AI
• Benefits
– Efficiency and Productivity
– Improved Decision Making
– Enhanced Personalization
– Innovation and New Capabilities
– Safety and Risk Management
• Risks
– Job Displacement
– Privacy and Security Concerns
– Bias and Fairness
– Lack of Transparency
– Ethical and Moral Issues
– Dependence and Reliability
AI vs ML vs RL vs DL

Deep Learning, I. Goodfellow, et. al.