CMSC 372 Artificial Intelligence Fall 2017 Administrivia Instructor: Deepak Kumar Lectures: Mon& Wed 10:10a to 11:30a Labs: Fridays 10:10a to 11:30a Pre requisites: CMSC B206 or H106 and CMSC B231 or permission of instructor Course web page: http://cs.brynmawr.edu/courses/cs372/fall2017/ 2 1
3 Machines with minds Decision making and problem solving Machines with actions Robots Thinking Like Acting Like Cognitive modeling/science Computational Psychology How does the brain work? Speech Understanding Turing Test Machines with logic laws of thought Logic Do the right thing Intelligent behavior in artifacts Thinking Rationally Acting rationally Reasoning Etc. Designing Intelligent Agents learning 4 2
Machines with minds Decision making and problem solving Thinking Like Cognitive modeling/science Computational Psychology How does the brain work? 5 Machines with minds Decision making and problem solving Cognitive Science Cognitive modeling/science Computational Psychology Thinking Like How does the brain work? Brain as an information processing system Requires theories of internal activities of the brain (level of abstraction? Knowledge or circuits?) How to validate? Predicting and testing behavior of human subjects (top down) Theories from neurological data (bottom up) Two fields: Cognitive Science & Cognitive Neuroscience 6 3
Machines with minds Decision making and problem solving Cognitive Science Cognitive modeling/science Computational Psychology Thinking Like How does the brain work? Brain as an information processing system Requires theories of internal activities of the brain (level of abstraction? Knowledge or circuits?) How to validate? Predicting and testing behavior of human subjects (top down) Theories from neurological data (bottom up) Two fields: Cognitive Science & Cognitive neuroscience Problem: Current theories do not explain anything resembling human level general intelligence. 7 Machines with actions Robots Acting Like Speech Understanding Turing Test 8 4
Machines with actions Robots Acting Like Speech Understanding Turing Test Alan Turing (1950) The Imitation Game 9 Machines with actions Robots The Turing Test Acting Like Speech Understanding Turing Test Alan Turing (1950) The Imitation Game 10 5
Machines with actions Speech Understanding Acting Like Alan Turing (1950) The Imitation Game Robots Turing Test The Turing Test Operational test for intelligent behavior Predicted that by 2000, a machine might have a 30% chance of fooling a lay person for 5 minutes. Anticipated all major arguments against AI in following 50 years Suggested major components of AI: knowledge, reasoning, language understanding, learning Major subfields of AI: language processing, knowledge representation & reasoning, machine learning, vision, robotics 11 Machines with actions Speech Understanding Acting Like Alan Turing (1950) The Imitation Game Robots Turing Test The Turing Test Operational test for intelligent behavior Predicted that by 2000, a machine might have a 30% chance of fooling a lay person for 5 minutes. Anticipated all major arguments against AI in following 50 years Suggested major components of AI: knowledge, reasoning, language understanding, learning Major subfields of AI: language processing, knowledge representation & reasoning, machine learning, vision, robotics Problem: Turing Test is not reproducible, constructive, or amenable to mathematical analysis 12 6
2015 Movie The Imitation Game 13 CAPTCHA? Completely Automated Public Turing test to tell Computers and Apart A reverse Turing Test? 14 7
Machines with logic laws of thought Logic Thinking Rationally Reasoning Etc. 15 Machines with logic laws of thought Logic Reasoning with Logic Reasoning Thinking Rationally Etc. Aristotle: What are correct arguments/thought processes? Formal Logics: Socrates is human. All humans are mortal. Therefore Socrates is mortal. Laws of thought govern the operation of the mind. 16 8
Machines with logic laws of thought Logic Reasoning with Logic Reasoning Thinking Rationally Etc. Aristotle: What are correct arguments/thought processes? Formal Logics: Socrates is human. All humans are mortal. Therefore Socrates is mortal. Laws of thought govern the operation of the mind. Problem: Not all intelligent behavior is mediated by logical deliberation. Not easy to formalize informal knowledge. E.g. Most students might be sleepy. 17 Do the right thing Intelligent behavior in artifacts Acting rationally Designing Intelligent Agents learning 18 9
Do the right thing Intelligent behavior in artifacts Rational Behavior Designing Intelligent Agents Acting rationally learning Do the right thing. That which is expected to maximize goal achievement, given available information. Doesn t necessarily involve thinking. E.g. blinking reflex. Any thinking there is, should be in service of rational action. Design Rational Agents. : 19 Do the right thing Intelligent behavior in artifacts Rational Behavior Designing Intelligent Agents Acting rationally learning Do the right thing. That which is expected to maximize goal achievement, given available information. Doesn t necessarily involve thinking. E.g. blinking reflex. Any thinking there is, should be in service of rational action. Design Rational Agents. : Problem: Computational limitations make perfect rationality unachievable. So, design best program for given computational resources. 20 10
Machines with minds Decision making and problem solving Machines with actions Robots Thinking Like Acting Like Cognitive modeling/science Computational Psychology learning Speech Understanding Turing Test Machines with logic laws of thought Logic Do the right thing Intelligent behavior in artifacts Thinking Rationally Acting rationally Reasoning Etc. Designing Intelligent Agents learning 21 Machines with minds Decision making and problem solving Machines with actions Robots Thinking Like Acting Like Cognitive modeling/science Computational Psychology learning Speech Understanding Turing Test Thought Processes Behaviors Machines with logic laws of thought Logic Do the right thing Intelligent behavior in artifacts Thinking Rationally Acting rationally Reasoning Etc. Designing Intelligent Agents learning 22 11
23 RoboCup Robot World Cup By the middle of the 21st century, a team of fully autonomous humanoid robot soccer players shall win a soccer game, complying with the official rules of FIFA, against the winner of the most recent World Cup. From: http://www.robocup.org/ 24 12
AI Beats FIFA 2052 Champs! What would it take to beat the FIFA World Cup 2052 champions? 25 AI Beats Jeopardy! Champion! 26 13
Watson: Jeopardy! https://youtu.be/p18edakuc1u February, 2011 27 Answer: Deepak Kumar 28 14
Watson/Jeopardy set at CHM 29 Is intelligence computable? 30 15
Is intelligence computable? Physical Symbol System Hypothesis a physical symbol system [such as a digital computer, for example] has the necessary and sufficient means for intelligent action. : Newell & Simon, 1976 31 Prehistory of AI Philosophy Mathematics Psychology Economics Linguistics Neuroscience Computer Science Control Theory Logic, methods of reasoning, mind as a physical system, foundations of learning, language, rationality Formal representation and proof, algorithms, computation, (un)decidability, (in)tractability, probability Adaptation, perception and motor control, experimental techniques Formal theories of rational decisions Knowledge representation, grammar Plastic physical substrate for mental activity Computer architectures, programming languages and paradigms (functional, declarative, OOP, etc.) Homeostatic systems, stability, optimal agent designs 32 16
Landmarks in AI History 1956 Birth of Artificial Intelligence Dartmouth Conference 1950 Alan Turing Computing Machinery &Intelligence 1943 McCulloch & Pitts Boolean Circuit model of brain 1952 69 Look Ma, no hands! GPS, Geometry Prob. Solver, Checkers, LISP 1940 1950 1960 1970 1980 1990 2000 2010 33 Landmarks in AI History 1969 79 Knowledge based systems DENDRAL, MYCIN, SHRDLU, PLANNER, CD, frames 1943 McCulloch & Pitts Boolean Circuit model of brain 1956 Birth of Artificial Intelligence Dartmouth Conference 1950 Alan Turing Computing Machinery &Intelligence 1952 69 Look Ma, no hands! GPS, Geometry Prob. Solver, Checkers, LISP 1969 Minsky & Papert: Perceptrons kills Neural Network Agenda 1966 ALPAC Report Machine Translation Killed 1965 Robinson Algorithm for logical reasoning 1962 Block et al Perceptron Convergence Theorem 1940 1950 1960 1970 1980 1990 2000 2010 34 17
Landmarks in AI History 1969 79 Knowledge based systems DENDRAL, MYCIN, SHRDLU, PLANNER, CD, frames 1943 McCulloch & Pitts Boolean Circuit model of brain 1956 Birth of Artificial Intelligence Dartmouth Conference 1950 Alan Turing Computing Machinery & Intelligence 1952 69 Look Ma, no hands! GPS, Geometry Prob. Solver, Checkers, LISP 1969 Minsky & Papert: Perceptrons kills Neural Network Agenda 1966 ALPAC Report Machine Translation Killed 1965 Robinson Algorithm for logical reasoning 1962 Block et al Perceptron Convergence Theorem 1980 AI becomes an industry Expert Systems boom 1976 Newell & Simon Physical Symbol System Hypothesis 1988 Resurgence of probability Nouvelle AI: Alife, GAs, soft computing HMMs, Bayes networks, data mining, ML 1985 95 Rebirth of Neural networks PDP, Connectionist models, Backprop 1990 AI Winter Expert Systems go bust 1940 1950 1960 1970 1980 1990 2000 2010 35 Landmarks in AI History 1969 79 Knowledge based systems DENDRAL, MYCIN, SHRDLU, PLANNER, CD, frames 1943 McCulloch & Pitts Boolean Circuit model of brain 1956 Birth of Artificial Intelligence Dartmouth Conference 1950 Alan Turing Computing Machinery & Intelligence 1952 69 Look Ma, no hands! GPS, Geometry Prob. Solver, Checkers, LISP 1969 Minsky & Papert: Perceptrons kills Neural Network Agenda 1966 ALPAC Report Machine Translation Killed 1965 Robinson Algorithm for logical reasoning 1962 Block et al Perceptron Convergence Theorem 1980 AI becomes an industry Expert Systems boom 1976 Newell & Simon Physical Symbol System Hypothesis 1988 Resurgence of probability Nouvelle AI: Alife, GAs, soft computing HMMs, Bayes networks, data mining, ML 1985 95 Rebirth of Neural networks PDP, Connectionist models, backprop 1990 AI Winter Expert Systems go bust 2001 AI Spring? HRI, data driven AI 1995 Agents everywhere! Robots, subsumption, human level AI 1940 1950 1960 1970 1980 1990 2000 2010 36 18
Landmarks in AI History 1969 79 Knowledge based systems DENDRAL, MYCIN, SHRDLU, PLANNER, CD, frames 1943 McCulloch & Pitts Boolean Circuit model of brain 1956 Birth of Artificial Intelligence Dartmouth Conference 1950 Alan Turing Computing Machinery & Intelligence 1952 69 Look Ma, no hands! GPS, Geometry Prob. Solver, Checkers, LISP 1969 Minsky & Papert: Perceptrons kills Neural Network Agenda 1966 ALPAC Report Machine Translation Killed 1965 Robinson Algorithm for logical reasoning 1962 Block et al Perceptron Convergence Theorem 1980 AI becomes an industry Expert Systems boom 1976 Newell & Simon Physical Symbol System Hypothesis 1988 Resurgence of probability Nouvelle AI: Alife, GAs, soft computing HMMs, Bayes networks, data mining, ML 1985 95 Rebirth of Neural networks PDP, Connectionist models, backprop 1990 AI Winter Expert Systems go bust 1995 Agents everywhere! Robots, subsumption, human level AI 2006 AI yields advances Self driven cars, MAPGEN, DEEP BLUE Home robots, Spam filters, etc. 2001 AI Spring? HRI, data driven AI Machine Learning make a lot of noise. Mostly driven by Big data and hardware advances. Goes commercial 2011 Big Data AI Watson, Deep Q&A Language translation 1940 1950 1960 1970 1980 1990 2000 2010 37 Agenda History, Foundations, Examples: Overview Intelligent Agents Problem Solving Using Classical Search Techniques Beyond Classical Search Adversarial Search & Game Playing Constraint Satisfaction Problems Knowledge Representation & Reasoning (KRR) First Order Logic & Inference Classical Planning Planning & Acting in the Real World Other topics depending upon time... 38 19
Acknowledgements Much of the content in this presentation is based on Chapter 1, Artificial Intelligence: A Modern Approach, by Russell & Norvig, Third Edition, Prentice Hall, 2010. This presentation is being made available by Deepak Kumar for any and all educational purposes. Please feel free to use, modify, or distribute. Powerpoint file(s) are available upon request by writing to dkumar@cs.brynmawr.edu Prepared in January 2015, updated September 2017. 39 20