Mirpur University of Science & Technology Mirpur

AJK, Department of Software Engineering

Lecture # 01

Artificial Intelligence

Engr Abdul Qadir Khan

[email protected]
1. Definitions of AI:
• Thinking Humanly:
• Described as an effort to make computers think with minds in a literal sense.
• Involves the study of mental faculties through computational models.
• Thinking Rationally:
• Involves the automation of activities associated with human thinking, such as
decision-making and problem-solving.
• Focuses on the study of computations enabling perception, reasoning, and action.
• Acting Humanly:
• Involves creating machines that perform functions requiring intelligence when
performed by people.
• Described as the study of the design of intelligent agents.
• Acting Rationally:
• Concerned with making computers do things at which, currently, people excel.
• Defined as the study of intelligent behavior in artifacts.
2. Dimensions of AI:
• Definitions are laid out along two dimensions: thought processes vs. behavior and fidelity
to human performance vs. rationality.
• Rationality is defined as doing the "right thing" given what is known.
3. Approaches to AI:
• Four approaches to AI: Thinking Humanly, Thinking Rationally, Acting Humanly, and
Acting Rationally.
• Historical development shows different people following different methods.
4. Thinking Humanly (Cognitive Modeling):
• Employs introspection, psychological experiments, and brain imaging to understand
human thought processes.
• Cognitive science combines AI models with experimental techniques from psychology to
construct theories of the human mind.
5. Acting Rationally (Rational Agent Approach):
• An agent is something that acts, and a rational agent acts to achieve the best outcome or
the best expected outcome.
 • Emphasizes more than correct inferences; includes autonomous operation, perception,
persistence, adaptation, and goal pursuit.
• The approach is more general and amenable to scientific development compared to
approaches based on human behavior or thought.
6. Achieving Rationality:
• Perfect rationality is the standard, but it's acknowledged that achieving it in complicated
environments is not feasible due to computational demands.
Foundations of Artificial Intelligence: A Historical Overview
In tracing the foundations of Artificial Intelligence (AI), we explore key disciplines that
contributed concepts, viewpoints, and techniques. The journey begins with fundamental
questions across various domains:
Philosophy:
• Key Questions:
• Can formal rules draw valid conclusions?
• How does the mind emerge from the physical brain?
• Where does knowledge originate, and how does it guide actions?
• Historical Contributors:
• Aristotle formulated laws governing the rational mind.
• Descartes introduced dualism, distinguishing mind and matter.
• Empiricism emerged, emphasizing knowledge from sensory experience.
Mathematics:
• Key Questions:
• What are the formal rules for valid conclusions?
• What can be computed?
• How to reason with uncertain information?
• Contributions:
• Formal logic, initiated by George Boole, laid the groundwork.
• Computation theory explored what could be algorithmically computed.
• Probability theory, from Cardano to Bayes, handled uncertain information.
Economics:
 • Key Questions:
• How to make decisions for maximum payoff?
• How to decide when others may not align?
• How to optimize decisions for future payoffs?
• Influence:
• Economics, treating economies as agents maximizing utility, contributed decision
theory.
• Game theory extended decision theory to interactions among multiple agents.
Neuroscience:
• Key Question:
• How do brains process information?
• Milestones:
• Advancements from Broca's study of brain localization to modern imaging
techniques.
• Neuroscience strives to unravel how the brain enables thought.
Psychology and Behaviorism:
• Key Questions:
• How do humans and animals think and act?
• Development:
• Behaviorism, led by Watson, rejected mental processes for observable behaviors.
• Cognitive science emerged, marrying psychology with computer modeling.
Computer Engineering:
• Key Question:
• How to build an efficient computer?
• Innovation:
• Invention of the modern digital electronic computer was pivotal for AI.
Control Theory and Cybernetics:
• Key Question:
• How can artifacts operate under their own control?
 • Pioneers:
• Cybernetics, led by Wiener, explored self-regulating systems and feedback
control.
Linguistics and Computational Linguistics:
• Key Questions:
• How does language work, and how can it be understood by machines?
• Intersection:
• Linguistics and AI intersected, giving rise to computational linguistics.
• Early challenges in language understanding spurred developments in knowledge
representation.
Example:
• Consider the evolution of AI in healthcare:
• Philosophy: Ethical considerations in AI-assisted medical decision-making.
• Mathematics: Algorithms for medical image analysis and predictive modeling.
• Economics: Optimization of resource allocation in healthcare systems.
• Neuroscience: Brain-inspired AI models for understanding and treating
neurological disorders.
• Psychology and Behaviorism: AI-driven mental health applications and therapy
bots.
• Computer Engineering: Hardware advancements for efficient medical AI
processing.
• Control Theory and Cybernetics: AI systems adapting responses based on
patient feedback.
• Linguistics and Computational Linguistics: Natural language interfaces for
medical chatbots.
1. The Gestation of Artificial Intelligence (1943–1955):
• Warren McCulloch and Walter Pitts (1943): Proposed a model of artificial neurons,
laying the groundwork for neural networks.
• Donald Hebb (1949): Introduced Hebbian learning, an influential model for modifying
connection strengths between neurons.
2. The Birth of Artificial Intelligence (1956):
 • Dartmouth Workshop (1956): John McCarthy, Marvin Minsky, Claude Shannon, and
others organized a workshop to explore the idea that every aspect of intelligence could be
precisely described, simulated, and improved through machines.
• Allen Newell and Herbert Simon: Developed the Logic Theorist, a program capable of
non-numerical thinking.
3. Knowledge-Based Systems: The Key to Power? (1969–1979):
• DENDRAL Program (1969): Ed Feigenbaum, Bruce Buchanan, and Joshua Lederberg
developed a knowledge-based system to infer molecular structures from mass
spectrometer data.
4. AI Becomes an Industry (1980–Present):
• Commercial Expert Systems: R1, the first successful commercial expert system,
operated at Digital Equipment Corporation, saving millions of dollars.
• Japanese Fifth Generation Project (1981): Japan initiated a 10-year plan to build
intelligent computers, leading to the formation of the Microelectronics and Computer
Technology Corporation (MCC) in the U.S.
• AI Industry Boom (1980–1988): AI became a billion-dollar industry, with numerous
companies working on expert systems, vision systems, and robotics.
5. The Return of Neural Networks (1986–Present):
• Back-Propagation Algorithm (Mid-1980s): Rediscovery and widespread use of the
back-propagation learning algorithm for neural networks.
• Connectionist Models: Considered as competitors to symbolic and logic-based AI
models.
6. The State of the Art:
• Robotic Vehicles: Autonomous vehicles like STANLEY and BOSS demonstrated
advanced capabilities in navigating complex terrains.
• Speech Recognition: Automated systems, like those used by United Airlines, enabled
natural language interaction.
• Autonomous Planning and Scheduling: NASA's Remote Agent program demonstrated
autonomous planning for spacecraft operations.
8. Natural Language Processing (NLP):
• Language Understanding: AI systems, like chatbots and virtual assistants, have
improved natural language understanding, enabling more sophisticated interactions with
users.
 • Language Translation: Advanced language translation models, such as those using
transformer architectures, have achieved remarkable accuracy.
9. Computer Vision:
• Object Recognition: Deep learning models have significantly improved object
recognition in images and videos, contributing to applications in autonomous vehicles,
security systems, and more.
• Facial Recognition: Facial recognition technology has become more prevalent in
security, authentication, and personalized user experiences.
10. Machine Learning and Deep Learning:
• Deep Learning Advances: Breakthroughs in deep learning, especially with
convolutional neural networks (CNNs) and recurrent neural networks (RNNs), have led
to improved performance in various tasks, from image recognition to natural language
processing.
• Transfer Learning: Techniques like transfer learning have enabled models to leverage
knowledge gained from one task to perform well in related tasks, reducing the need for
extensive training data.
11. Reinforcement Learning:
• Game Playing: Reinforcement learning algorithms have demonstrated exceptional
performance in playing complex games, including board games like Go and video games.
12. AI in Healthcare:
• Medical Diagnosis: AI systems, particularly deep learning models, are making strides in
medical image analysis for diagnosing conditions such as cancer from radiological
images.
• Drug Discovery: AI is being used to accelerate drug discovery processes by predicting
potential drug candidates and their effects.
13. Autonomous Systems:
• Drones and Robotics: AI-powered drones and robots are increasingly used for tasks
such as surveillance, delivery, and exploration.
• Autonomous Vehicles: Ongoing advancements in self-driving car technology have the
potential to transform transportation.