01 introduction


Published on

Dr. Ali abdalla Lecture

Published in: Education, Technology, Spiritual
1 Like
  • Be the first to comment

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

01 introduction

  1. 1. Artificial Intelligence TextbookRussell and Norvig: Chap. 1 and 2
  2. 2. Today’s Agenda  Introduction to AI  Overview of the course  Search Problems
  3. 3. Various “Definitions” of AI AI is the reproduction of the methods of human reasoning or intuition AI uses computational models to simulate intelligent (human) behavior and processes AI is the study of mental faculties through the use computational methods
  4. 4. Various “Definitions” of AI Intelligent behavior Computer Humans
  5. 5. What is AI? Discipline that systematizes and automates reasoning processes to create machines that: Act like humans Act rationally Think like humans Think rationally
  6. 6. Act like humans Act rationally Think like humans Think rationally The goal of AI is to create computer systems that perform tasks regarded as requiring intelligence when done by humans  AI Methodology: Take a task at which people are better, e.g.: • Prove a theorem • Play chess • Plan a surgical operation • Diagnose a disease • Navigate in a building and build a computer system that does it automatically But do we want to duplicate human imperfections?
  7. 7. Act like humans Act rationally Think like humans Think rationally Here, how the computer performs tasks does matter The reasoning steps are important  Ability to manipulate symbolic knowledge (lemmas, concepts, …)
  8. 8. Discourse on the Method, byDescartes (1598-1650)“If there were machines which bore a resemblance toour bodies and imitated our actions as closely aspossible for all practical purposes, we should still havetwo very certain means of recognizing that they werenot real men. The first is that they could never usewords, or put together signs, as we do in order todeclare our thoughts to others… Secondly, even thoughsome machines might do some things as well as we dothem, or perhaps even better, they would inevitablyfail in others, which would reveal that they are actingnot from understanding, …”
  9. 9. Turing Test Test proposed by Alan Turing in 1950 The computer is asked questions by a human interrogator. It passes the test if the interrogator cannot tell whether the responses come from a person Required capabilities: natural language processing, knowledge representation, automated reasoning, learning,... No physical interaction
  10. 10. Can Machines Act/ThinkIntelligently? Yes, if intelligence is narrowly defined as information processing In fact, AI has made impressive achievements showing that tasks initially assumed to require intelligence can be automated Probably not, if intelligence is not separated from the rest of “human nature”
  11. 11. Some Big Open Questions AI (especially, the “rational agent” approach) assumes that intelligent behaviors are based on information processing? Is this a valid assumption? If yes, can the human brain machinery solve problems that are inherently intractable for computers? In a human being, where is the interface between “intelligence” and the rest of “human nature”, e.g.:  How does intelligence relate to emotions felt?  What does it mean for a human to “feel” that he/she understands something? Is this interface critical to intelligence? Can there exist a general theory of intelligence independent of human beings?
  12. 12. In the serie “Star Trek, The NewGeneration” the most impressive featureof the robots Data and his clone is nottheir ability to solve complex problems,but how they blend human-like reasoningwith other key aspects of human beings(especially, self-consciousness, fear ofdying, distinction between right andwrong, prefer music, etc…)
  13. 13. AI has made impressive achievements showing thattasks initially assumed to require intelligence can beautomatedAI can be seen as contributing to building aninformation processing model of human beings, justas Biochemistry contributes to building a model ofhuman beings based on bio-molecular interactionsFrom two different perspectives, both try to explainhow a human being operates. Both also explore waysto avoid human imperfections (in Biochemistry, byengineering new proteins and drug molecules; in AI,by designing rational reasoning methods) . Both tryto push the limits of their foundational assumption
  14. 14. Place of AIin Computer ScienceUnique approach:Take a task thought to require intelligence, andautomate it AI-specific representations and algorithms:search algorithms, formal logic, machine learning,etc... AI ways of analyzing these representations and algorithms Relations with other areas: automatic control,operational research, game theory
  15. 15. Main Areas of AI Search, especially heuristic search (puzzles, Agent Perception games) Robotics Knowledge representation (including formal logic) Reasoning Planning Search Reasoning with Learning uncertainty, including probabilistic reasoning Knowledge Constraint rep. Learning Planning satisfaction Agent architectures Robotics and perception Natural language Natural ... Expert processing language Systems
  16. 16. Bits of History 1956: The name “Artificial Intelligence” is coined (John McCarthy) 60’s: Search and games, formal logic and theorem proving 70’s: Robotics, perception, knowledge representation, expert systems 80’s: More expert systems, AI becomes an industry 90’s: Rational agents, probabilistic reasoning, machine learning 00’s: Systems integrating many AI methods
  17. 17. Some Achievements Computers have won over world champions in several games, including Checkers, Othello, and Chess, but still do not do well in Go AI techniques are used in many systems and applications, e.g.: formal calculus, video games, route planning, logistics planning, pharmaceutical drug design, medical diagnosis, hardware and software trouble- shooting, speech recognition, road traffic monitoring, facial recognition, medical image analysis, part inspection, etc... In fact, there are few complex computer systems, if any, that use no AI method Some industries (automobile, electronics) are highly robotized, while other robots perform brain and heart surgery, are rolling on Mars, fly autonomously, …, but home robots remain mostly a thing of the future
  18. 18. Required textbook:S. Russell and P. Norvig. Artificial Intelligence:A Modern Approach. Second edition, PrenticeHall, 2003[you can have it at the midterm and finalexams]
  19. 19. Tentative ScheduleDate Topic Out Russell & Norvig textbook10/08 Introduction / Search problems Chap. 1 and 210/15 Search problems Chap. 3, Sect. 3.1–2 + 3.610/22 Blind search Chap. 3, Sect. 3.3–510/29 Heuristic search Chap. 4, Sect. 4.1–311/05 Action planning HW1 Chap. 11, 11.1–411/12 Adversarial Search Chap. 611/19 Multiple Agents Environments Chap. 17 Sect. 17.6-17.711/26 Knowledge Representation HW2 Chap. 8, Chap. 10. Sect. 10.312/03 Midterm12/10 Inductive learning Chap. 18.1, 18.3/ Introduction to uncertainty Chap. 13/ Non-deterministic uncertainty HW3 Chap. 12/ Deciding under probabilistic uncertainty Chap. 17/ Bayesian nets Chap. 14/ Review & Conclusion
  20. 20. HWs, midterm, Exam• 3 HWs• 3 weeks due delay• Submit by e-mail. No late days• Midterm (open book)• Final (open book)• Final Grade: 10% for each HW, 30% Midterm, 40% Final Exam