Lecture 1 Introduction To Artificial Intelligence PDF

Lecture 1 Introduction To Artificial Intelligence Pro. Osama Abdel Raouf Announcements & Course Materials Required Resources: The prescribed textbooks for the course are: - Stuart J. Russell and Peter Norvig, "Artificial Intelligence: A Modern Approach," 3rd Edition (2010), by Pe...

Lecture 1 Introduction To Artificial Intelligence Pro. Osama Abdel Raouf Announcements & Course Materials Required Resources: The prescribed textbooks for the course are: - Stuart J. Russell and Peter Norvig, "Artificial Intelligence: A Modern Approach," 3rd Edition (2010), by Pearson Education Inc. -George F. Luger, "Artificial Intelligence: Structures and strategies for complex problem solving, " 6th Edition (2008), Pearson Education Limited. Additional Textbooks: - Wolfgang Ertel, "Introduction to Artificial Intelligence," 2nd Edition (2017) - Miroslav Kubat, "An Introduction to Machine Learning," 2nd Edition (2017) 2 Student assessments + Assessment Grades 10% (5% Attendance) + (5% Project) 20% Quizzes (2 quizzes each 10%) 20% Midterm Exam 50% Final Exam Class Project o Work to be done in groups of 2-3 students, more details will be announced later. o Final report will include the analysis of your data, including code and visual results. o The students will be required to perform a ~5-10 mins demo of their project. 4 Outline & Learning Objectives.. of this course 5 Learning Objectives of this Course What you'll learn?  This course gives a basic introduction to Artificial Intelligence (AI) and Machine Learning (ML).  Students receive an introduction to philosophical fundamental problems and ethical questions related to ML/AI, as well as the field's history.  Then, We will study the core topics of knowledge representation, reasoning, and learning.  Later topics will include and introduction to machine learning, & probabilistic reasoning, in addition to applications such as robotics, computer vision, and natural language processing. 6 Learning Objectives of this Course. Continued.. What you'll learn?  The course covers basic supervised classification (e.g., Artificial Neural Networks), as well as unsupervised learning (Clustering), optimization (Evolutionary Algorithms and other search methods), and tentatively regression and reinforcement learning.  practical Through an algorithmic approach, the understanding students beingtaught, are given of thea methods implementati in particular several of the through making ons of methods.their own 7 Topics Covered The main contents of the course are:  Main approaches to AI & Learning  Task environment  Performance measures  Intelligent Agents  Knowledge Representation  Problem solving by searching  Uninformed search  Informed search  Beyond classical search: Evolutionary Algorithms  Machine learning  Supervised learning versus Unsupervised learning  Decision trees  Neural networks  Support vector machines  Cross validation 8 Lecture 1: An introduction to Artificial Intelligence [AI] 1.1 What is Intelligence 1.3 AI as the Study & Design  Some Foundations of Artificial of Intelligent Agents Intelligence  Systems that Think like  What is Intelligence? Humans  What is Artificial Intelligence?  Systems that Think Rationally 1.2 Systems that Act Like Humans  Challenges to Systems that Think Rationally  Systems that Act Like Humans  Systems that Act Rationally  Turing Test (the Imitation Game)?  AI as the Study & Design of Intelligent Agents  Total Turing Test?  Intelligent Agents in the  The Chinese Room Argument World  Strong Vs. Weak AI   4 An Introductio n to Artifici al Intellige  This lecture covers the following chapters: Chapter 1 (Introduction) from Stuart J. Russell and Peter Norvig, "Artificial Resourc Intelligence: A Modern Approach," esfor Third Edition thislect (2010), by Pearson Education Inc. ure.. AND.. Chapter 1 (AI: History and Applications) from George F. Luger, "Artificial Intelligence: 5 SOME FOUNDATIONS OF ARTIFICIAL INTELLIGENCE Philosophy Economics Can formal rules be used to How should we make decisions draw valid so as conclusions? to maximize payoff? How does the mind arise from a How should we do this when physical brain? others Where does knowledge come may not go along? from? How should we do this when How does knowledge lead to the action? payoff may be far in the future? Mathematics Computer Engin What are the formal rules to eering draw valid How can we build an efficient conclusions? computer? What can be computed? Control theory and How do we reason with cybernetics uncertain How can artefacts operate information? under their 6 own control?  What is Intelligence? Intelligence: Judgment, otherwise called “good sense,” “practical sense,” “initiative,” the faculty of adapting one's self to circumstances.. auto-critique ~ Alfred Binet (July 8, 1857 – October 18, 1911) was a French psychologist who invented the first practical intelligence test (An intelligence quotient (IQ); a total score derived from one of several standardized tests designed to assess human intelligence) “.. the resultant of the process of acquiring, storing in memory, retrieving, combining, comparing, and using in new contexts information and conceptual skills.” ~Lloyd G. Humphreys (December 12, 1913 – September 7, 2003) was an American psychologist “.. the capacity to learn and solve problems..” (Webster’s dictionary) in particular, the ability to solve novel problems the ability to act rationally the ability to act like humans 7  What is Artificial Inteligence? John Mc Carthy*, Stanford University What is artificial intelligence? 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, but AI does not have to confine itself to methods that are biologically observable; “.. The goal of AI is to develop machines that behave as though they were intelligent...” More in: http://www-formal.stanford.edu/jmc/whatisai/node1.html * John McCarthy (September 4, 1927 – October 24, 2011) was an American computer scientist & cognitive scientist. McCarthy was one of the founders of the discipline of artificial intelligence. He coined the term "artificial intelligence" (AI). 8  What is Artificial Inteligence? by E n c y c l o p e d i a B r i t a n n i c a ( 1991 ) ".. AI is the ability of digital computers or computer-controlled robots to solve problems that are normally associated with the higher intellectual processing capabilities of humans.“ by E l a i n e R i c h.. Artificial Int elligence. McGraw-Hill, 1983 ".. Artificial Intelligence is the study of how to make computers do things at which, at the moment, people are 15  What is Artificial Inteligence? Four Main Approaches that have been followed, each by different people with different methods. Thinki Acting ng Huma Huma nly nly Thinki Acting ng Ration Ration ally ally 16  What is Artificial Inteligence? Systems that act like humans Systems that think rationally “The study of how to make “The study of mental faculties computers do things at which, at the through the use of computational moment, people are better” (Rich models” (Charniack and McDermott, and Knight, 1991) 1985). “The art of creating machines that “The study of the computations that perform functions that require make it possible to perceive, reason, intelligence when performed by and act.” (Winston, 1992) people.” (Kurzweil, 1990) Systems that think like humans Systems that act rationally “The automation of activities that we associate with human thinking, such “AI.. is concerned with intelligent as decision making, problem solving, behavior in artifacts (Nilsson, 1998) learning” (Bellman, 1978) “Computational Intelligence is the “The exciting new effort to make study of the design of intelligent computers think … machines with agents.” (Poole et al., 1998) minds, in the full and literal sense.” (Haugeland, 1985) 17  What is Artificial Inteligence? Systems that act like humans Systems that think rationally “The study of how to make “The study of mental faculties computers do things at which, at the through the use of computational moment, people are better” (Rich models” (Charniack and McDermott, and Knight, 1991) 1985). “The art of creating machines that “The study of the computations that perform functions that require make it possible to perceive, reason, intelligence when performed by and act.” (Winston, 1992) people.” (Kurzweil, 1990) Systems that think like humans Systems that act rationally “The automation of activities that we associate with human thinking, such “AI.. is concerned with intelligent as decision making, problem solving, behavior in artifacts (Nilsson, 1998) learning” (Bellman, 1978) “Computational Intelligence is the “The exciting new effort to make study of the design of intelligent computers think … machines with agents.” (Poole et al., 1998) minds, in the full and literal sense.” (Haugeland, 1985) Systems that Act Like Humans … ?!  Systems that Act Like Humans TuringTest;theImitationGame… In Turing’s (1950) paper “Computing machinery and intelligence”: ♦Can machines think ? ≡ (identical to) Can machines behave intelligently? ♦Operational test for intelligent behavior: the Imitation HUMAN Game HUMAN ? INTERROGATOR AI SYSTEM  Systems that Act Like Humans Turing Test; The Imitation Game… Turing test (1950): Can a human interrogator tell whether (written) responses to her (written) questions come from a human or a machine? Natural Language Processing Knowledge Representation Automated Reasoning Machine Learning Total Turing Test (extended to include physical aspects of human behavior): Computer Vision 21 Robotic  Total Turing Test? But why do we want an intelligent system to act like a human? - Because for many tasks, humans are still the Gold Standard. 22  Baby BabyXis a project (by Auckland's Institute Laboratory for Animate Bioengineering X! a virtual animated Technologies) to makebaby that learns and reacts like a human baby. It uses the Total Turing Test? computer's cameras for "seeing" and microphones to "listen" as the inputs. The computer uses AI algorithms for BabyX's "learning" and interpretation of the inputs (voice and image) to understand the situation. The result is a virtual toddler that can learn to read, recognize objects and "understand." The output is the baby's face that can "speak" and express its mood by 23 facial expressions (such as smiling).  Baby Reinforcement learning..? It is a machine learning training method based X! on rewarding desired behaviors and/or punishing undesired ones. Total Turing Test? Affective Computing..? it describes computing that is in some way connected to emotion ( a.k.a. emotional artificial intelligence). It is the study and development of systems and devices that can recognize, interpret, process, and simulate human affects (feelings, 24  Systems that Act Like Humans The Chinese Room Argument John Rogers Searle (born July 31, 1932) is an American philosopher “Searle's thought experiment begins with this hypothetical premise: suppose that artificial intelligence research has succeeded in constructing a computer that behaves as if it understands Chinese. It takes Chinese characters as input and, by following the instructions of a computer program, produces other Chinese characters, which it presents as output. Suppose, says Searle, that this computer performs its task so convincingly that it comfortably passes the Turing test: it convinces a human Chinese speaker that the program is itself a live Chinese speaker. To all of the questions that the person asks, it makes appropriate responses, such that any Chinese speaker would be convinced that they are talking to another Chinese-speaking human being.” The question Searle wants to answer is this: does the machine literally "understand" Chinese? Or is it merely simulating the ability to understand Chinese? Searle calls the first position "strong2 AI" 2  Systems that Act Like Humans The Chinese Room Argument (Continued) Searle then supposes that he is in a closed room and has a book with an English version of the computer program, along with sufficient paper, pencils, erasers, and filing cabinets. Searle could receive Chinese characters through a slot in the door, process them according to the program's instructions, and produce Chinese characters as output. If the computer had passed the Turing test this way, it follows, says Searle, that he would do so as well, simply by running the program manually. Searle asserts that there is no essential difference between the roles of the computer and himself in the experiment. Each simply follows a program, step-by- step, producing a behaviour which is then interpreted as demonstrating intelligent conversation. However, Searle would not be able to understand the conversation. Searle he argues concludes that that "strong without "understanding" (or 2 3  Systems that Act Like Humans The Chinese Room Argument (Continued) If person inside does a great job of answering questions, can we say s/he understands? Even if (s)he is only blindly following rules? (Obviously, the ‘person inside’ is acting like an AI program) 27  Systems that Act Like Humans The Chinese Room Argument (Continued) Strong vs. Weak AI Hypotheses? - WEAK AI Hypothesis; We can accurately simulate animal / human intelligence in a computer. - STRONG AI Hypothesis; We can create algorithms that are intelligent ( Consciousne ss ?.. Self- Awareness ?. Do you. Free-will ?) Sonn remember the y, robot from / science-fiction 200 the action film "I, 4 Robot"? 28  Systems that Act Like Humans Strong Vs. Weak AI.. Where are we? Artificial Super Intelligence Machine Intelligence Artificial Narrow Intelligence Source: https:// 2 www.upwork.com/hiring/for-clients/artificial-intelligence-and-natural-language-processing-in-big-d 6  Lecture 2: An introduction to Artificial Intelligence [AI] 1.1 What is Intelligence  Some Foundations of AI 1.3 AI as the Study & Design  What is Intelligence? of Intelligent Agents  What is Artificial Intelligence?  Systems that Think like Humans 1.2 Systems that Act Like  Systems that Think Humans Rationally  Systems that Act Like Humans  Challenges to Systems that  Turing Test (the Imitation Think Rationally Game)?  Systems that Act Rationally  Total Turing Test?  The Chinese Room Argument  Strong Vs. Weak AI  What is Artificial Inteligence? Systems that act like humans Systems that think rationally “The study of how to make “The study of mental faculties computers do things at which, at the through the use of computational moment, people are better” (Rich models” (Charniack and McDermott, and Knight, 1991) 1985). “The art of creating machines that “The study of the computations that perform functions that require make it possible to perceive, reason, intelligence when performed by and act.” (Winston, 1992) people.” (Kurzweil, 1990) Systems that think like humans Systems that act rationally “The automation of activities that we associate with human thinking, such “AI.. is concerned with intelligent as decision making, problem solving, behavior in artifacts (Nilsson, 1998) learning” (Bellman, 1978) “Computational Intelligence is the “The exciting new effort to make study of the design of intelligent computers think … machines with agents.” (Poole et al., 1998) minds, in the full and literal sense.” (Haugeland, 1985)  Systems that Think Like Humans Need to study the brain as an information processing machine, … in other words … it is based on cognitive modeling approach. Steps to develop an AI that think like human: 1. Use Computational Models to Understand the Actual Workings of Human Mind 2. Devise/Choose a sufficiently precise theory of the mind. 3. Express it as a computer program. 4. Check match between program and human behavior (actions and timing) on similar tasks. Tight connections with Cognitive Science & Neuroscience. Finally building systems that can model the cognitive capability of the human. Simulating the way the human think, conclude, reason with. How can convert these to computations and based on these computations we can build these AI systems. Examples: Chatbots (GPT-3 or GPT-4) or Virtual personal assistants (Siri, Alexa, google assistant)  What is Artificial Inteligence? Systems that act like humans Systems that think rationally “The study of how to make “The study of mental faculties computers do things at which, at the through the use of computational moment, people are better” (Rich models” (Charniack and McDermott, and Knight, 1991) 1985). “The art of creating machines that “The study of the computations that perform functions that require make it possible to perceive, reason, intelligence when performed by and act.” (Winston, 1992) people.” (Kurzweil, 1990) Systems that think like humans Systems that act rationally “The automation of activities that we associate with human thinking, such “AI.. is concerned with intelligent as decision making, problem solving, behavior in artifacts (Nilsson, 1998) learning” (Bellman, 1978) “Computational Intelligence is the “The exciting new effort to make study of the design of intelligent computers think … machines with agents.” (Poole et al., 1998) minds, in the full and literal sense.” (Haugeland, 1985)  Systems that Think Rationaly Logic: That is; patterns of arguments (fact) that always yields correct conclusions when supplied with correct premises. It is based on laws of thought approach or logic. “.. Socrates is a man; all men are mortal; therefore Socrates is mortal.” Steps to perform systems that think rationally: 1. Try to build computational frameworks based on logic, that is, describe a problem in formal logical notation and apply general deduction (conclusion) procedures to solve it. 2. Then use these frameworks to build intelligent systems that think rationally. Semantic Tight connections with (Propositional Logic) Networks. and (Logic Programming). More advanced logic-based representations: 3 3  What is Artificial Inteligence? Systems that act like humans Systems that think rationally “The study of how to make “The study of mental faculties computers do things at which, at the through the use of computational moment, people are better” (Rich models” (Charniack and McDermott, and Knight, 1991) 1985). “The art of creating machines that “The study of the computations that perform functions that require make it possible to perceive, reason, intelligence when performed by and act.” (Winston, 1992) people.” (Kurzweil, 1990) Systems that think like humans Systems that act rationally “The automation of activities that we associate with human thinking, such “AI.. is concerned with intelligent as decision making, problem solving, behavior in artifacts (Nilsson, 1998) learning” (Bellman, 1978) “Computational Intelligence is the “The exciting new effort to make study of the design of intelligent computers think … machines with agents.” (Poole et al., 1998) minds, in the full and literal sense.” (Haugeland, 1985)  Systems that Act Rationaly Why..? Acting rationally refers to the process of making decisions and taking actions that lead to desirable outcomes or goals, based on available information and knowledge. (information or knowledge obtained using any approach) Aim to build rational agents to achieve goals. These agents can imitate the human actions or use the cognitive modeling of the human or use the law of thought to reach the goal. (I don’t care). But more is needed for rational behavior, e.g. How to behave when there is no provably correct thing to do (i.e. reasoning under uncertainty). Examples: Autonomous Robotics: uses sensor data, algorithms to see the environment to optimize the actions and achieve the objectives. (manufacturing, healthcare,.etc) Traffic Management systems: in smart cities (use data from cameras, sensors, and traffic flow model to optimize the traffic signal THANK YOU