01 what is AI

8/6/2019 01 what is AI

http://slidepdf.com/reader/full/01-what-is-ai 1/49

Artificial Intelligence:introduction

Stefano De Luca

8/6/2019 01 what is AI


13/10/2008 2AI

Pag.

Practical stuff

Stefano De Luca Mail:

– [email protected]– [email protected]

Receive: after lesson, by appointment

Textbook:– S. Russell and P. Norvig Artificial

Intelligence: A Modern ApproachPrentice Hall, 2003, Second Edition

– In Italian: I n t e l l igenza ar t i f i c ia le . Un app roccio m ode rno , 1° vol., PearsonEducation Italia,2005, €40

– Exam:

– written at end and oral– If you ask for: intermediate written exam

mailto:[email protected]




8/6/2019 01 what is AI


13/10/2008 3AI

Pag.

Course overview

What is AI. Intelligent agents.

Knowledge representation RDF, OWL Semantic web Logic and Reasoning

Problem solving. Planning Uncertain knowledge and reasoning. Neural Network

Genetic Algorithms Industrial case histories

8/6/2019 01 what is AI

http://slidepdf.com/reader/full/01-what-is-ai 4/4913/10/2008 4AI

Pag.

Outline

What is AI A brief history

The State of the art (see book)

8/6/2019 01 what is AI

http://slidepdf.com/reader/full/01-what-is-ai 5/4913/10/2008 5AI

Pag.

What is Artificial Intelligence

Creative extension of philosophy:– Understand and BUILD intelligent entities

Origin after WWII

Highly interdisciplinary

Currently consist of huge variety of subfields– This course will discuss some of them

8/6/2019 01 what is AI


13/10/2008 6AI

Pag.

Old AI…

8/6/2019 01 what is AI


13/10/2008 7AI

Pag.

Now: Smart Music System

The Bose uMusic

system uses artificialintelligence to learnthe listening habits andpreferences of its users.

Load your CDs into thedigital music deliverysystem, it can holdthousands of songs, and

it will learn yourlistening preferencesand prioritize yourmusic collection.

8/6/2019 01 what is AI


13/10/2008 8AI

Pag.

Lastfm.it, Pandora.com

8/6/2019 01 what is AI


13/10/2008 9AI

Pag.

Control systems

8/6/2019 01 what is AI


13/10/2008 10AI

Pag.

What is Artificial Intelligence

Different definitions due to

different criteria– Two dimensions:

– Thought processes/reasoning vs. behavior/action

– Success according to human standards vs. successaccording to an ideal concept of intelligence: rationality.

Systems that think like humans Systems that think rationally

Systems that act like humans Systems that act rationally

8/6/2019 01 what is AI


13/10/2008 11AI

Pag.

Systems that act like humans

When does a system behave intelligently?– Turing (1950) Computing Machinery and Intelligence

– Operational test of intelligence: imitation game

– Test still relevant now, yet might be the wrong question.

– Requires the collaboration of major components of AI:knowledge, reasoning, language understanding, learning, …

8/6/2019 01 what is AI


13/10/2008 12AI

Pag.

Systems that act like humans

Andrew Hodges.Alan Turing, the enigma(Storia di un Enigma,Bollati Boringhieri)

Problem with Turing test: not reproducible,constructive or amenable to mathematical analysis.

8/6/2019 01 what is AI


13/10/2008 13AI

Pag.

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 notexplain anything resembling humanintelligence.

– Three fields share a principal direction.

8/6/2019 01 what is AI


13/10/2008 14AI

Pag.

Systems that think like humans

Some references;– Daniel C. Dennet.

Consciousness explained.

– M. Posner (edt.)Foundations of cognitivescience

– Francisco J. Varela et al.The Embodied Mind

– J.-P. Dupuy. Themechanization of the mind

8/6/2019 01 what is AI


13/10/2008 15AI

Pag.

Systems that think rationally

Capturing the laws of thought

– Aristotle: What are ‘correct’ argument andthought 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:– Not all intelligence is mediated by logic behavior

– What is the purpose of thinking? What thought shouldone have?

8/6/2019 01 what is AI


13/10/2008 16AI

Pag.

Systems that think rationally

A reference;– Ivan Bratko, Prologprogramming for

artificial intelligence.

8/6/2019 01 what is AI


13/10/2008 17AI

Pag.

Systems that act rationally

Rational behavior: “doing the right

thing” – The “Right thing” is that what is expected to

maximize goal achievement given the

available information.

Can include thinking, yet in service

of rational action.–Action without thinking: e.g. reflexes.

8/6/2019 01 what is AI


13/10/2008 18AI

Pag.


Two advantages over previous approaches:– More general than law of thoughts approach

– More amenable to scientific development.

Yet rationality is only applicable in idealenvironments.

Moreover rationality is not a very good modelof reality.

8/6/2019 01 what is AI


13/10/2008 19AI

Pag.


Somereferences

8/6/2019 01 what is AI


13/10/2008 20AI

Pag.

Rational agents

An agent is an entity that perceives and acts

This course is about designing rationalagents

– An agent is a function from percept histories to actions:

– For any given class of environments and task we seek theagent (or class of agents) with the best performance.

– Problem: computational limitations make perfect rationalityunachievable.

f : P*→ A

8/6/2019 01 what is AI


13/10/2008 21AI

Pag.

Foundations of AI

Different fields have contributed to AI in the

form of ideas,viewpoints and techniques.– Philosophy: Logic, reasoning, mind as a physical 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 agentdesign.

8/6/2019 01 what is AI


13/10/2008 22AI

Pag.

A brief history

What happened after WWII?

– 1943: Warren Mc Culloch and Walter Pitts: a model of artificial boolean neurons to perform computations.– First steps toward connectionist computation and learning

(Hebbian learning).

– Marvin Minsky and Dann Edmonds (1951) constructed the firstneural network computer

– 1950: Alan Turing’s “Computing Machinery and

Intelligence” – First complete vision of AI.

– Idea of Genetic Algorithms

8/6/2019 01 what is AI


13/10/2008 23AI

Pag.

A brief history (2)

The birth of (the term) AI (1956)

– Darmouth Workshop bringing together top minds onautomata 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

– Arthur Samuel (1952-)investigated game playing (checkers ) with greatsuccess.

– John McCarthy(1958-) :– Inventor of Lisp (second-oldest high-level language)

– Logic oriented, Advice Taker (separation between knowledge and reasoning)

8/6/2019 01 what is AI


13/10/2008 24AI

Pag.

A brief history (3)

The birth of AI (1956)

– Great expectations continued ..– Marvin Minsky (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 systems lacked scalability.– Combinatorial explosion in search.

– Fundamental limitations on techniques and representations.– Minsky and Papert (1969) Perceptrons.

8/6/2019 01 what is AI


13/10/2008 25AI

Pag.

A brief history (4)

AI revival through knowledge-basedsystems (1969-1970)– General-purpose vs. domain specific

– 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, …

8/6/2019 01 what is AI


13/10/2008 26AI

Pag.

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); back-propagation

8/6/2019 01 what is AI


13/10/2008 27AI

Pag.

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”

8/6/2019 01 what is AI


13/10/2008 28AI

Pag.

The Hype Cycle of Emerging Technologies(Gartner 2005)

8/6/2019 01 what is AI


13/10/2008 29AI

Pag.

Major research areas (Applications)

Natural Language Understanding Image, Speech and patternrecognition

Uncertainty Modeling

Expert systems

Virtual Reality

…..

8/6/2019 01 what is AI


Pag.

Program as Representation of world

Symbol as basic element of representation– atom, property, relationship

Symbolic Expression as method of

combination LISP for Symbolic programming PROLOG for logic programming

Object-Oriented Concept

Symbolic Programming

8/6/2019 01 what is AI


Pag.

Knowledge Representation

What kind of Knowledge needed forProblem solving ?

Structure of knowledge ?

– declarative vs procedural

Representation techniques ?– explicit vs (implicit + inference)

– logic, frame, object-oriented, semantic net, script

Knowledge acquisition and update

8/6/2019 01 what is AI


Pag.

Search Theory

An Optimization method

Analyze alternative cases and select one Cope with Exponential complexity, NP

classes– Try likely one first (Heuristic Search)– Utilize local information (Hill Climbing Method)– Optimal solution vs good solution

Genetic Algorithm, Simulated Annealing– Stochastic search

8/6/2019 01 what is AI


Pag.

Automated Reasoning

Qualitative Reasoning– Utilization of qualitative knowledge such as

Non-monotonic Reasoning

– Ostrich flys ? Plausible Reasoning

– Information fusion under uncertainty

Case-based Reasoning– Utilization of Experience

8/6/2019 01 what is AI


Pag.

Performance improvement by experience– How much of knowledge required to start learning ?– Method of acquiring new knowledge and merging it to existing

knowledge-base– Role of teacher

– Role of examples and experience Parameter Adjustment Inductive learning

Computational Learning Theory– Quality of generalization capability in terms of Training data

Used in Practice such as Data Mining

Machine Learning

8/6/2019 01 what is AI


13/10/2008 35AI

Pag.

Data Mining

Data Information / knowledge DecisionMaking

Knowledge extraction for decision making

N l N k

8/6/2019 01 what is AI


Pag.

Computational model of Neurons–Power comes from Connection of

simple processing element -

connectionism

Σ

X1

X2

Xn

.

.

.

w1

w2

wn

F( X 1 , X 2 , …, X n)

Neural Network

N l N t k

8/6/2019 01 what is AI


13/10/2008 37AI

Pag.

learning = link weigh adjustment– Error-back-propagation : supervised learning

– Any Functional Mapping is learnable

Strong at Sensory Data Processing– Symbolic Grounding

Old Horse on the race again– Massive parallelism, graceful degradation

Neural Network

N l N t k Cl ifi

8/6/2019 01 what is AI


13/10/2008 38AI

Pag.

Neural Network Classifier

Job(1/0)

age

Salary

#mouth

Debt

good

medium

bad

Inputlayer

Hiddenlayer

Outputlayer

G ti Al ith

8/6/2019 01 what is AI


Pag.

Genetic Algorithms

Computational model of lifeevolution

Stochastic optimization technique

– Initial chromosome creation– New generations are made (cross over,

mutation)

– survival of the fittest Base of artificial life research

– study evolution of life, by simulation

AI S St (Pl i )

8/6/2019 01 what is AI


13/10/2008 40AI

Pag.

AI Success Story (Planning)

MARVEL (Schwuttke, 1992)

– Real-time Space shuttle Missionplanning

Berth assignment (KAL, 1997)

Unmanned Vehicle– Ground and air

Pathfinder Rover, 1996 Asimo – a walking robot

Autonomous Land Vehicle

8/6/2019 01 what is AI


13/10/2008 41AI

Pag.

Autonomous Land Vehicle(DARPA’s GrandChallenge contest)

AI Success Story : Medical expert

8/6/2019 01 what is AI


13/10/2008 42AI

Pag.

AI Success Story : Medical expert

systems

Antibiotics & InfectiousDiseasesCancer

Chest pain Dentistry

DermatologyDrugs & ToxicologyEmergency

EpilepsyFamily Practice

Genetics Geriatrics

Programs listed by Special Field

GynecologyImaging Analysis

Internal Medicine

Intensive Care

Laboratory Systems

Orthopedics

Pediatrics

Pulmonology & VentilationSurgery & Post-OperativeCare

Trauma Management

Pattern Recognition Applications

8/6/2019 01 what is AI


13/10/2008 43AI

Pag.

Pattern Recognition Applications

Handwriting and document recognition

– forms, postal mail, historic documents– PDA pen recognition

Signature, biometrics (finger, face, iris, etc.)

Gesture, facial expression– As a Human computer intertraction

EEG, EKG, X-rayTrafic monitoring, Remote Sensing

Smart Weapon – guided missile, targethoming

Handwriting Understanding

8/6/2019 01 what is AI


13/10/2008 44AI

Pag.

Handwriting Understanding

전자 펜으로 수식 입력 수식 인식

Decision Support Systems (DSS)

8/6/2019 01 what is AI


13/10/2008 45AI

Pag.

Decision Support Systems (DSS)

Intelligent Transportation Systems

8/6/2019 01 what is AI


13/10/2008 46AI

Pag.

Intelligent Transportation Systems

Future of AI

8/6/2019 01 what is AI


13/10/2008 47AI

Pag.

Future of AI

Making AI Easy to use– Easy-to-use Expert system building tools

– Web auto translation system

– Recognition-based Interface Packages

Integrated Paradigm– Symbolic Processing + Neural Processing

AI in everywhere, AI in nowhere– AI embedded in all products

– Ubiquitous Computing, Pervasive Computing

Other courses at the UniRoma2

8/6/2019 01 what is AI


13/10/2008 48AI

Pag.

Other courses at the UniRoma2

AI does not end here …–Sistemi di agenti (Specialistica)

–Cooperazione di Agenti informatici(Specialistica)

–…

Some references

8/6/2019 01 what is AI


13/10/2008 49AI

Pag.

Some references

Understanding Intelligence by Rolf Pfeifer and Christian Scheier.

Artificial Intelligence: Structuresand Strategies for ComplexProblem-solving by George Luger.

Computation and Intelligence:Collective readings edited byGeorge Luger.

Paradigms of Artificial IntelligenceProgramming: Case Studies inCommon Lisp by Peter Norvig.

01 what is AI

Documents

Transcript of 01 what is AI