Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Technological singularity wikipedia , lookup
Incomplete Nature wikipedia , lookup
Expert system wikipedia , lookup
Embodied cognitive science wikipedia , lookup
Knowledge representation and reasoning wikipedia , lookup
Intelligence explosion wikipedia , lookup
Philosophy of artificial intelligence wikipedia , lookup
Existential risk from artificial general intelligence wikipedia , lookup
HIT3002: Introduction to Artificial Intelligence Subject Overview Staff information Bao Vo (Lecturer, Convenor) Email: [email protected] Phone: (9214) 4756 Office: EN504 Li Minyi (Tutor) Email: [email protected] Swinburne University of Technology 1 Subject Roadmap Introduction Intelligent agents Search Knowledge representation and reasoning Planning Probabilistic reasoning and Bayesian networks Adaptation and learning Meeting time Lectures: When: Monday 11.30 – 13.30 Where: EN203 Tutorials: Starting Week 2 (March 7th) Group 1: Monday 13.30 – 14.30 (Room: EN406) Group 2: Monday 9.30 – 10.30 (Room: EN202) Consultations: When: Monday 15.30 – 17.00 Where: EN504 Swinburne University of Technology 2 Subject Assessment (Provisional) 2 Assignments – 20% and 20% Progress on assignment must be shown in tutorials Involves problem solving Exam – 60% Focus will be on conceptual understanding of all main topics covered in the lectures and the work of the tutorials Tutorials will help in facing the exam with confidence! Exam is E i a hhurdle! dl ! I.e., I you mustt gett att lleastt 50% on the th final fi l exam to pass the subject. Subject Information Textbook Russell, S.J., Norvig, P., Artificial Intelligence: A Modern Approach, 3rd edition, Prentice-Hall, 2010. [AIMA] References B. Coppin “Artificial Intelligence Illuminated” Jones and Bartlett Publishers, 2004 Nilsson “Artificial Intelligence: A New Synthesis” Morgan Kaufman Pub. 1998 Core Aims Understand fundamental concepts of Artificial Intelligence (AI) and generic problem solving techniques Apply advanced algorithms and data structures to solve common problems Design simple software that implements AI concepts. Swinburne University of Technology 3 Lectures Focus is on theory High-level conceptual discussion of the algorithms Assignment discussion (10-15 minutes) Q & A of topics covered so far (10-15 minutes) Tutorials Aim is to ensure that all examinable aspects (theory) are fully covered in these sessions Discussion on all fundamental topics in the subject Some simple problems that are to be solved using AI techniques Assignment progress monitoring Discussion on potential exam questions, sample exam p p paper Swinburne University of Technology 4 HIT3002: Introduction to Artificial Intelligence Introduction Why Artificial Intelligence (AI) It’s not just a good topic for science fiction There are successful stories of using AI: Expert systems: makes 100’s of millions of dollars per year (for the mining and banking industries) AI techniques in search engines (specifically targeted by Google) AI techniques in automotive industry: BMW’s iDrive, Holden’s AI stabilisation system Fuzzy-logic controllers can be found in washing machines, dish washers, camera autofocus, air conditioners, car automatic transmissions, etc. Swinburne University of Technology 5 Why Artificial Intelligence (AI) Last year HIT3002 students’ success stories: A student received the Faculty Summer ’08 Vacation Scholarship to work on a pproject j at Thales Groupp ((Australia)) Using AI techniques to develop solutions for Air Traffic Control Management (ATM) systems Offered an internship by Thales to continue to work on the project while studying full-time in his last year at Swinburne Another student received the Faculty Summer ’08 Vacation Scholarship to work on a project at Swinburne developing nextgeneration ti social i l networks t k Another student received the Faculty Summer ’08 Vacation Scholarship to work on a project for cloud computing Why study AI It provides the core knowledge of computer science You’ll learn to analyse problems and learn about t h i techniques/algorithms / l ith tto solve l real-world l ld problems bl It paves the way to understanding various sorts of intelligence (in both humans and machines) It is also fun (and different to most other subjects) Swinburne University of Technology 6 What is intelligence Faculty of understanding Capacity to know or apprehend Available ability as measured by intelligence tests or social criteria Ability to use knowledge in new situations or problems Ability to learn Ability to plan, to foresee problems Ability to use symbols and relationships Abilit to think abstractly, Ability abstractl to work ork towards to ards goals Ability to act/perform certain functions … you name it What is Artificial Intelligence Different definitions due to different criteria Two dimensions: Thought processes/reasoning vs. behavior/action Success according to human standards vs. success according to an ideal concept of intelligence: rationality. Systems y that think like humans Systems y that think rationallyy Systems that act like humans Swinburne University of Technology Systems that act rationally 7 Systems that act like humans AI is the art of creating machines that perform functions that require intelligence when performed by humans Methodology: Take an intellectual task at which people are better and make a computer do it Turingg test •Prove a theorem •Play chess •Plan a surgical operation •Diagnose a disease •Navigate in a building Systems that act like humans When does a system behave intelligently? Turing (1950) Computing Machinery and Intelligence Operational O ti l ttestt off iintelligence: t lli iimitation it ti game Test still relevant now, yet might be the wrong question. Requires the collaboration of major components of AI: knowledge, reasoning, language understanding, learning, … Swinburne University of Technology 8 Systems that think like humans How do humans think? Requires scientific theories of internal brain activities (cognitive model): Level of abstraction? (knowledge or circuitry?) Validation? Predicting and testing human behavior Identification from neurological data Cognitive Science vs. Cognitive neuroscience. Both approaches are now distinct from AI Share that the available theories do not explain anything resembling human intelligence. Three fields share a principal direction. Systems that think like humans Some references; Daniel C. Dennet. Consciousness explained. M. Posner (edt.) Foundations of cognitive science Francisco J. Varela et al. The Embodied Mind J.-P. Dupuy. The mechanization of the mind Swinburne University of Technology 9 Systems that think rationally Capturing the laws of thought Aristotle: What are ‘correct’ argument and thought processes? Correctness depends on irrefutability of reasoning processes. This study initiated the field of logic. The logicist tradition in AI hopes to create intelligent systems using logic programming. Problems: P bl Not all intelligence is mediated by logic behavior What is the purpose of thinking? What thought should one have? Systems that think rationally A reference; Ivan Bratko, Prolog programming for artificial intelligence. Swinburne University of Technology 10 Systems that act rationally Rational behavior: “doing the right thing” The “Right Right thing” thing is the course of action that is expected to maximize goal achievement given the available information. Can include thinking, yet action. in service of rational Action without thinking: e.g. reflexes. Systems that act rationally Two advantages over previous approaches: More M generall than th law l off thoughts th ht approachh More amenable to scientific development. Yet rationality is only applicable in ideal environments. Moreover rationality is not a very good model of reality. Swinburne University of Technology 11 Systems that act rationally A reference: Rolf Pfeifer and Christian Scheier. Understanding Intelligence. MIT Press, 1999. Foundations of AI Different fields have contributed to AI in the form of ideas,viewpoints and techniques. Philosophy: Logic, Logic reasoning reasoning, mind as a physical system system, foundations of learning, language and rationality. Mathematics: Formal representation and proof algorithms, computation, (un)decidability, (in)tractability, probability. Psychology: adaptation, phenomena of perception and motor control. Economics: formal theory of rational decisions, game theory. Linguistics: knowledge represetatio, grammar. Neuroscience: physical substrate for mental activities. Control theory: homeostatic systems, stability, optimal agent design. Swinburne University of Technology 12 A brief history What happened after WWII? 1943: 1943 Warren W Mc M Culloch C ll h andd W Walter lt Pitt Pitts: a model d l off artificial boolean neurons to perform computations. First steps toward connectionist computation and learning (Hebbian learning). Marvin Minsky and Dann Edmonds (1951) constructed the first neural network computer 1950: Alan Turing’s “Computing Machinery and Intelligence” First complete vision of AI. A brief history (2) The birth of AI (1956) Darmouth Workshop bringing together top minds on automata theory, neural nets and the study of intelligence. Allen Newell and Herbert Simon: The logic theorist (first nonnumerical thinking program used for theorem proving) For the next 20 years the field was dominated by these participants. Great expectations (1952-1969) Newell and Simon introduced the General Problem Solver. Imitation of human problem-solving problem solving Arthur Samuel (1952-)investigated game playing (checkers ) with great success. John McCarthy(1958-) : Inventor of Lisp (second-oldest high-level language) Logic oriented, Advice Taker (separation between knowledge and reasoning) Swinburne University of Technology 13 A brief history (3) The birth of AI (1956) Great expectations continued .. Marvin Minskyy ((1958 -)) Introduction of microworlds that appear to require intelligence to solve: e.g. blocks-world. Anti-logic orientation, society of the mind. Collapse in AI research (1966 - 1973) Progress was slower than expected. Unrealistic predictions. Some S systems t lacked l k d scalability. l bilit Combinatorial explosion in search. Fundamental limitations on techniques and representations. Minsky and Papert (1969) Perceptrons. A brief history (4) AI revival through knowledge-based systems (19691970) General-purpose G l vs. ddomain i specific ifi E.g. the DENDRAL project (Buchanan et al. 1969) First successful knowledge intensive system. Expert systems MYCIN to diagnose blood infections (Feigenbaum et al.) Introduction of uncertainty in reasoning. Increase in knowledge representation research. Logic, frames, semantic nets, … Swinburne University of Technology 14 A brief history (5) AI becomes an industry (1980 - present) R1 at DEC (McDermott, 1982) Fifth generation project in Japan (1981) American response … Puts an end to the AI winter. Connectionist revival (1986 - present) Parallel distributed processing (RumelHart and McClelland, 1986); backprop. A brief history (6) AI becomes a science (1987 - present) Neats vs. scruffies. In speech recognition: hidden markov models In neural networks In uncertain reasoning and expert systems: Bayesian network formalism … The emergence of intelligent agents (1995 - present) The whole agent problem: “How does an agent act/behave embedded in real environments with continuous sensory inputs” Swinburne University of Technology 15 State of the art Deep Blue defeated the reigning world chess champion Garry Kasparov (1997) Proved a mathematical conjecture (Robbins conjecture) unsolved for decades “No hands” across America (driving autonomously 98% of the time from Pittsburgh to San Diego) During the 1991 Gulf War, US force deployed an AI logistics planning and scheduling program that involved up to 50,000 vehicles, cargo, and people NASA’s on-board autonomous planning program controlled the scheduling of operations for a spacecraft Proverb solves crossword puzzles better than most humans Summary Different people think of AI differently. Two important questions to ask are: Are you concerned with ith thinking thi ki or behavior? b h i ? Do D you wantt to t model d l hhumans or work from an ideal standard? In this course, we adopt the view that intelligence is concerned mainly with rational action. Ideally, y, an intelligent g agent g takes the best ppossible action in a situation. We will study the problem of building agents that are intelligent in this sense. Swinburne University of Technology 16