صفحه 1:
Petree 7
Computer Engineering Department
Artificial Intelligence:
۳۰۱۶۰ Present and Future
Parham Moradi
A
ار http://math-cs.aut.ac.ir/~moradi
صفحه 2:
آیا ميتوان ماشيني طراحي کرد که در بزرگراههاي تهران به صورت
اتوماتیک رانندگي کند؟
آیا ميتوان رباتي طراحي کرد که بدون کمک انسان قضاياي جدید رياضي را
اثبات کند؟
آیا ميتوان رباتي طراحي کرد که بتواند در مورد موضوع خاصي یک داستان
یا مقاله بنویسد؟
آیا ميتوان یک ربات پزشک طراحي کرد ؟
آيا ميتوان یک ماشین مترجم ساخت صحبتهاي کسي را که به زبان فارسي
با یک انگليسي زبان گفتگو ميکند را به صورت اتوماتیک ترجمه کند؟
و
صفحه 3:
What is Al
Foundation of Al
A brief history
Al Research Areas
Application of Al
Future of Al
صفحه 4:
What is Artificial Intelligence
Intelligence: “ability to learn, understand and
think” (Oxford dictionary)
the study of the capacity of machines to
simulate intelligent human behaviour
Al is the study of how to make computers make
things which at the moment people do better.
صفحه 5:
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 that think Systems that think
Systems that act Systems that act
صفحه 6:
Systems that like humans
When does a system behave intelligently?
- Turing (1950) Computing Machinery and Intelligence
- Operational test of intelligence: imitation game
3 |
و ۴:5
- Test still relevant now, yet might be the wrong question.
- Requires the collaboration of major components of Al:
knowledge, reasoning, language understanding, learning, ...
صفحه 7:
Systems that like humans
Turing Test
1990: Loebner Prize established. Grand Prize of
$100,000 and a Gold Medal for the first computer
whose responses are indistinguishable from a
human.
(Solid 18 carat, not gold-plated like the Olympic
"Gold" medals)
صفحه 8:
Systems that
a اله الل شا 2 a 3 ع اه
Srourmy shop. sep ل oss Scniae Read Home _Searen Wlatecape
MER" whats Rataiow ]7 مه سم و aE Bootranks 2 Locator: |
[ Potenbers که P comectins J aniowna f snanspaate f vroiace
aS Create your own robot personal
fa
CHAT ROBOT fea
People Are Saying to ALICE Now:
‘What can I call you? "1
eI 1
~ [ Repiy] لقع
صفحه 9:
Systems that 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 Al
Share that the available theories do not explain
anything resembling human intelligence.
- Three fields share a principal direction.
صفحه 10:
Systems that like humans
صفحه 11:
Systems that like humans
WE MIND Tame beware
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
CONSCIOUSNESS
EXPLAINED و
صفحه 12:
Systems that 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 Al 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 should one
have?
صفحه 13:
Systems that
A reference; PROLOG
- Ivan Bratko, Prolog ا
programming for
artificial
intelligence.
صفحه 14:
Systems that 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.
صفحه 15:
Systems that rationally
Two advantages over previous approaches:
- More general than law of thoughts approach
- More amenable to scientific development.
Yet rationality is only applicable in idea/
environments.
Moreover rationality is not a very good model
of reality.
صفحه 16:
Rational agents
An agent is an entity that perceives and acts
This course is about designing rational
agents
- An agent is a function from percept histories to actions:
f:PK5 A
- For any given class of environments and task we seek the
agent (or class of agents) with the best performance.
- Problem: computational limitations make perfect rationality
unachievable.
صفحه 17:
Intelligence — is the ability to make primitive judgment by
logical arguments, which comes from past experience.
Al —is the development of technique which can be used
to reproduce this ability in computers and other machine.
صفحه 18:
A brief history
Foundation of Al
Al Research Areas
Application of Al
Future of Al
صفحه 19:
Foundation of Al
A brief history
Al Research Areas
Application of Al
Future of Al
صفحه 20:
Build systems that display intelligent
behaviour i.e ” Smart systems”
Technological perspective
Methods
Knowledge based methods
‘Behavioral methods
‘Subsymbolic methods
Connected via emperical method
Foundation of Al
Studie of intelligent systems
related to computational processes
Scientific perspective
ulilding on ia
صفحه 21:
Foundations of Al
ilosophy (423 BC - present):
- Logic, methods of reasoning.
Mind as a physical system.
ll- Foundations of learning, language, and rationality.
S (c.800 - present):
- Formal representation and proof.
11- Algorithms, computation, decidability, tractability.
il- Probability.
صفحه 22:
Foundations of Al
Jy (1879 - present):
- Adaptation
il- Phenomena of perception and motor
{il- Experimental techniques.
(1957 - present):
- Knowledge representation.
il- Grammar.
صفحه 23:
A brief history
Al Research Areas
Application of Al
Future of Al
صفحه 24:
A brief history
- 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 first
neural network computer
- 1950: Alan Turing’s “Computing Machinery and
Intelligence”
- First complete vision of Al.
- Idea of Genetic Algorithms
صفحه 25:
A brief history (2)
The birth of (the term) Al (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
~ 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)
صفحه 26:
A brief history (3)
The birth of Al (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 Al 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.
صفحه 27:
A brief history (4)
Al revival through knowledge-based
systems (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, ...
صفحه 28:
A brief history (5)
Al becomes an industry (1980 - present)
- R1 at DEC (McDermott, 1982)
- Fifth generation project in Japan (1981)
- American response ...
Puts an end to the Al winter.
Connectionist revival (1986 - present)
- Parallel distributed processing (RumelHart and McClelland,
1986); back-propagation
صفحه 29:
A brief history (6)
Al 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”
صفحه 30:
Al Research Areas
Application of Al
Future of Al
صفحه 31:
Major research areas (appiications)
Knowledge Representation
Natural Language Understanding
Image, Speech and pattern recognition
Uncertainty Modeling
Expert systems
Web Mining
صفحه 32:
Knowledge Representation
What kind of Knowledge needed for
Problem solving ?
Structure of knowledge ?
- declarative vs procedural
Representation techniques ?
- explicit vs (implicit + inference)
- logic, frame, object-oriented, semantic net, script
Knowledge acquisition and update
صفحه 33:
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
صفحه 34:
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
صفحه 35:
Machine Learning
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
صفحه 36:
Data Mining
Knowledge extraction for decision
صفحه 37:
Neural Network
omputational model o
Neurons
- Power comes from Connection of
simple processing element -
connectionism
xX, wa
Se ۳۳
% ههه ) FX, Xe or XD
3
x,
صفحه 38:
Neural Network
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
صفحه 39:
Neural Network Classifier
Job(1/0) On
وود 9>
Salary ——)<
#mouth — -@4 ب
Debt —— @f 2
صفحه 40:
Genetic Algorithms
Computational model of life
evolution
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
صفحه 41:
Web Mining
Analysing Web Information
Web 1.0
- Web Usage Mining
- Web Content Mining
- Web Structure Mining
Web 2.0
-Semantic Web
-Web Services
-Social Networks (Wikis-Web Logs-...)
صفحه 42:
Application of Al
Future of Al
صفحه 43:
Autonomous Land Vehicle
(DARPA’s GrandChallenge contest)
صفحه 44:
Autonomous Land Vehicle
DARPA’s GrandChallenge contest)
GPS Receiver
‘OP Postion
Inter Vehicle Signaling like Veticle Signaling
Mp updates and
esther feed
Fader ed for
وب
object cetecion
5-7 Wireless Internet
صفحه 45:
Automatic Speech Processing
صفحه 46:
Game Playing
ry Kasparov Vs Deep Blue
صفحه 47:
Game Playing
Computer won world هرا اك of chess
(S
صفحه 48:
Medical expert systems
rounnvs listed by Gpevial Piekt
Antibiotics & Infectious Gynecology
Diseases ۱ Anal
Caricer maging Analysis
Chest pain Internal Medicine
Dentistry Intensive Care
Dermatology Laboratory Systems
Drugs & Orthopedics
Toxicology ae
Emergency Pediatrics
Epilepsy Pulmonology & Ventilation
Family Practice Surgery & Post-Operative
Genetics Care
Geriatrics Trauma Management
صفحه 49:
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-ray
Trafic monitoring, Remote Sensing
Smart Weapon - guided missile, target
homing
صفحه 50:
Handwriting Understanding
اس سس اد ممه معد 2
sires na wee
سه وق ۳
۶ عم asso
soos دا
سه م i”
لمعه ونه pious
۰
إلنق اتن أعج سمه 0
صفحه 51:
Decision Support Systems (DSS)
صفحه 52:
Intelligent Transportation Systems
صفحه 53:
Smart Music System
The Bose uMusic
system uses artificial
intelligence to learn
the listening habits and
preferences of its users.
Load your CDs into the
digital music delivery
system, it can hold
thousands of songs, and
it will learn your
listening preferences
and prioritize your
music collection.
صفحه 54:
Pandora.com
22 ۳
5
موس 6 0 ۲
yr more music that I'l like?
ا وت [] سس ۲ موز موی Tenet ٩ © ]مد و
صفحه 55:
Control systems
صفحه 56:
Outline
Future of Al
صفحه 57:
Future of Al
هي يا Slut a 4 Forsight b Futurology ابداعات؛
نوآوريها و ادامه روندها و پيشبيني وقایع علمي در آینده ميپردازد
ميپردا دازد
صفحه 58:
Future of Al
اين فیلد در دنیا با کلمات كليدي زیر نیز شناخته ميشود :
وهزبا5 ۴۲۲۵5 (در محافلدلنشگاهیل
*ونعه0۲؟ 5۲۵۲691 (در محافلدلنشگاهی
futurology
futuristics
futures thinking
futuring
(awil, 6 5) futuribles
prospectiva (در لیتالا و آمریکایلا نی
صفحه 59:
Future of Al
واژه 50751 براي اولين بار در سال 9060 در سایت خبري BBC
busi ۲4۰63۰۷۷6115 براي تاسیس دانشکدهاي در اين زمینه جهت پيشبيني آینده
تحقيقاتي دانشگاهي coast dls PO به کار گرفته شد
. واژه /
براي اولین بار توسط یک دانشمند آلماني به نام Ossip K. Flechtheim »
سال 1960 به کار برده شد.
Fu كه معني آن» مطالعات در مورد آینده یا آيندهنگري است؛
در سال (QOD براي اولین بار در ایالت متحده آمریکا رشته دانشگاهي آینده
پژوهي تاسیس شد
صفحه 60:
Future of Al
به كساني که در فیلد آینده پژوهي مطالعه و تحقیق ميکنند اصطلاحا " آینده نگر "
Futurist & گفته ميشود
آینده نگر ها در مورد مسیر حرکت ale تحقیق مي کنند
. مطالعات برروي آيندهپژوهي از سال 19660 به بعد اغاز شد و افرادي چون
Herman Kahn, Olaf Helmer, Bertrand de Jouvenel,
Dennis Gabor, Oliver Markley, Burt Nanus, and
86۱۱ ۷۷606۱۱ از پایهگذاران اين علم بودند
صفحه 61:
Future of Al
بعضي ديكر از آينده نكر هاء نتيجه مطالعات و بيزوهش هاي خود را در زمينه آينده نكرم
صورت فیلمنامه در مي آوردند و فیلم هاي تخيلي a Ae! L (Science Fiction) cde
کردند
يكي از معروفترین آنها آرتور سي کلارک (13۲16 6 Gul (Arthure که اولین فیلم او با
این مضمون در سال 960 با عنوان "2001 : 00۷556۷ 5066 ۸ " به نمایش در
صفحه 62:
Future of Al
تعدادي از اين نوع فیلم ها در زیر لیست شده است :
A Space Odyssey - Arthur C. Clarke - 1968 :2001 [1]
Odyssey Two - Arthur C. Clarke - 1982 :2010[2]
Odyssey Three - Arthur C. Clarke - 1982 :2061 [3]
The Final Odyssey, the Leonov mission is said :3001 [4]
to have taken place in the 2040s - Arthur C. Clarke -
1997
The Year We Make Contact - 1984 :2010[5]
صفحه 63:
Future of Al
پیش بيني هاي مجله lu GD 2» The Futurist آینده
بیشتر تصمیمات» توسط موجوديتهاي غیر انساني اتخاذ (رباتها» سيستمهاي خبره و
...) خواهد شد
شبكهاي از تيمهاي الکترونيکي» رباتهاي هوشمند در حوزههاي مختلف She
سیاست» بهداشت » آموزش و عیره تصمیم گيري خواهند کرد. دلیل آن پیشرفت
تكنولوژي است و باعث کاهش خطاي انساني در تصمیم گيريها ميشود.
صفحه 64:
0
-١ كاربرد خانكى
بيه | 604
ب سر
55 04 =
ASIMO Abio Paero
یک ربات خانگی
ریات انساننمایی که قابلیت
راه وفتن دار
صفحه 65:
Future of Al
9 كاديودهاى يزشعى
‘Ri-Man 1
ريات تكن نان ريات لمداديار
ربات در تاروشانده Da Vinei
رباتی که عمل جراحی انجام
Es =
صفحه 66:
Future of Al
كاربردهاي نظامي
از سال 00000 به بعد سازمانهاي نظامي آمریکا پروژه بزرگي براي تولید
رباتهاي نظامي را شروع کردند و از رباتها در جنگ با عراق نیز توسط
آمریکا استفاده شد.
در 00یا 000 سال آیند سازمانهاي نظامي آمریکا تفنگ و ادوات کشنده را كنار
خواهند گذاشت و ادوات جنگي غیر کشنده را تولید خواهند کرد که اين ادوات
جنگي توسط رباتها استفاده خواهد شد.
صفحه 67:
Future of Al
كاربردهاي نظامي
از سال 00000 به بعد سازمانهاي نظامي آمریکا پروژه بزرگي براي تولید
رباتهاي نظامي را شروع کردند و از رباتها در جنگ با عراق نیز توسط
آمریکا استفاده شد.
در 00 یا 00 سال آیند سازمانهاي نظامي آمریکا تفنگ و ادوات کشنده را کنار
خواهند گذاشت و ادوات جنگي غیر کشنده را تولید خواهند کرد که اين ادوات
جنگي توسط رباتها استفاده خواهد شد.
تر جنگها در coal
او سط it jp طر نا صلحء sb hic
صفحه 68:
Future of Al
کاربردهاي نظامي
Machine-gun equipped | bate field extraction Big Dog
robot
صفحه 69:
Future of Al
صفحه 70:
Future of Al
رباتهاي کاوشگر فضا
صفحه 71:
Future of Al
پیش بينيهاي دیگر در مورد رباتها
رباتهاي كشاورزي - از سال 00000 تاسال COUP
رباتهايي که از سالمندان مراقبت ميکنند - از سال ۸00 تا سال 600
رباتهايي که عمل جراحي جزيي انجام ميدهند - تا سال 600072
رباتهايي که كارهاي خانه را انجام ميدهند - از سال 200072 تا سال 0009
نانو رباتها
یک سوم آتشنشانها » ربات خواهند بود- تا سال 60006
جنگها کاملا بوسیله
009
باتهاي سرباز خود مختار انجام ميشود - تا سال
صفحه 72:
Future of Al
پیش بينيهاي دیگر در مورد رباتها
رباتهاي كشاورزي - از سال 00000 تاسال COUP
رباتهايي که از سالمندان مراقبت ميکنند - از سال ۸00 تا سال 600
رباتهايي که عمل جراحي جزيي انجام ميدهند - تا سال 600072
رباتهايي که كارهاي خانه را انجام ميدهند - از سال 200072 تا سال 0009
نانو رباتها
یک سوم آتشنشانها » ربات خواهند بود- تا سال 60006
جنگها کاملا بوسیله
009
باتهاي سرباز خود مختار انجام ميشود - تا سال
صفحه 73:
Future of Al
پیش بينيها در حوزههاي دیگر
- جستجوي وب بر اساس محتوي صفحات وب
- آنالیز حجم زياد داده در وب
مثال :
- لیست عكسهايي که در آنها ماشین و درخت وجود دارد
- استفاده از وب به عنوان منبع دانش : آیا ميتوان شرح حال بیمار را گفت و
یک سیستم خبره » با استفاده از مستندات وب» دارو تجویز کند؟
Google
صفحه 74:
Future of Al
پیش بينيها در حوزههاي دیگر
تر جمه ماشيني و پردازش سیگنال
- سازمانهاي نظامي آمریکا سرگرم طراحي یک ماشین آي سي هستند که
بتواند در داخل موبایل ها قرار گیرد و ترجمه گفتار را بر اساس 49 زبان
زنده دنیا انجام دهد
- ترجمه اتوماتیک صفحات وب Google
صفحه 75:
Future of Al
هوش شناختي (Cognitive Science)
- مطالعه برروي نحوه رفتار و تفکر انسان
صفحه 76:
با تشکر از توجه شما