Intr
od
ucti
on
Natural intelligence
Artificial intelligence (AI)
History of AI
Characteristics of AI system
AI tasks
AI roots and applications
natural
WHAT A BRAIN CAN DO?
Recognize a ball
Plan
Search solution
Understand speech/language
Diagnose disease
Predict weather
Solve maths problem
Learn
Make decision
Commonsense reasoning……………………………
natural
AI is ….an ‘imitation’ of human intelligence into a
machine
naturalartificial
natural
Not consistent
Cannot be copiedor transferred
High cost
Difficult to bedocumented
Creative
Wider focus
artificial
Consistent
Can be copiedor transferred
Low cost
Can bedocumented
Limited focus
DEFIN
ITIO
N“science and engineering of making
intelligent machines”
“science of making computers DO things
that require intelligence if they are done by humans”
“science of making a computer
REASONin a manner similar to humans”
John McCarthy
Marvin Minsky
Durkin Jackson
DIM
EN
SIO
N Systemthinks
like human
System
acts like human
System
thinks rationally
System acts
rationally
DIM
EN
SIO
N Cognitive Science perspective(thinking process)
Engineering perspective(behavior)
DIM
EN
SIO
NSystem thinks like
human
System acts
like human
System thinks
rationally
System acts
rationally
DIM
EN
SIO
N
Compare system
against real concept of intelligence
Compare system against human
HIS
TO
RY
1st AI winter
LISP
Dartmouth workshop
Period of enthusiasm
Turing test
1st NN computer built
1st commercialization
Rebirth
2nd AI winter
Re-emergence
1966-19731958
1956
1952-19621950
1950
1980
1969-1979
1988
2000s
AI
EV
OLU
TIO
NEmbedded application
Multiple integration
Domain knowledge
General method
Naïve solution
2000 onwards
1990s
1980s
1970s
1960s
This game ends when the interrogator made his guess
Turing machine pass the test if the interrogator fails to recognize with
whom (or with what) he is communicating.
TURING MACHINE
TU
RIN
G
TES
T
1 - Knowledge processing
on(a,b)
on(b,c)
on(c,table)
clear(a)
ABC
CH
AR
AC
TER
IST
IC The birds fly
D VN
VPNP
S
the birds fly
s(np(det(the),noun(birds)),vp(v(fly)))
2 - Focus on heuristic
4 - Possess inference ability
3 - Symbolic processing
5 - Learns
CH
AR
AC
TER
IST
IC
Mundane
Formal
ExpertAI
TA
SK
S
“We may not have full-
functioning robots that cater our
every need, but AI is
embedded in our everyday
lives.”- Computer
WorldAP
PLIC
ATIO
N
“Once tools get far
enough out of the lab,
they’re no longer AI,
just common computer science.”
- Prof. George LugerA
PP
LIC
ATIO
N
Household gadgets
• washing machine• rice cooker• microwave
• vacuum cleaner
AP
PLIC
ATIO
N
Autonomous vehicle
Hi, my name is Stanley
AP
PLIC
ATIO
N
Search engine crawlers
NewsFinder
FlipDog
NetFlix
(natural language processing technology)A
PP
LIC
ATIO
N
Automobile safety system
self-parking
intelligent
cruisecontrol
lanedeparturewarning
AP
PLIC
ATIO
N
Voice response system(IVR)
Speech recognition Natural language understanding
AP
PLIC
ATIO
N
Chatter botJenn
Alex
Eva
Captain Kirk
Alice
George
(natural language processing technology)A
PP
LIC
ATIO
N
Resource allocation problem
Optimizing airport schedule
(genetic algorithm technology)A
PP
LIC
ATIO
N
Credit card fraud detectionFalcon
(neural network technology)
Pap smear screeningFocalPoint
(expert system technology)
AP
PLIC
ATIO
N