Mohammad Raihanul

Islam

Lecturer
Stamford
University
• Movies are all
dominated by
intelligent agent
• Intelligent agent are
created because of the
development of AI
Not an easy definition !!!
Two main dimension
First one is related to human
•System think like human
•Machine with minds
•System act like human
•Creating machine that perform function
require intelligence
Second one measure against rationality
•System thinks rationally
•Using computer models
•Act rationally
•Designing intelligent agents
Natural language processing
Knowledge representation
Automated reasoning
Machine learning
Computer vision
Robotics
How we think?
Need to get inside the actual workings of
human mind
Two way to do this
•Trying to catch our own thoughts as the go
•Psychological experiments
Cognitive Science
•Computer models from AI
•Experimental models from psychology
Thinking rightly
Logic
•Laws of thought govern the operation of
mind
Agent
•Something that acts
Rational Agent
•Act to gain maximum benefit
Making correct decision sometimes proved
to rational
Sometime it is not true!!
Sometimes no probable correct thing to do
but something must be done
Has at least two benefits
•Correct inferences are several possibilities
for achieving rationality
•More suitable than human thought based
methods
Prize competition for driverless vehicles
funded by the Defense Advanced Research
Projects Agency
2004 Grand Challenge:
•150 mile route in Mojave Desert
• Carnegie Mellon University's Red Team and car
Sandstorm (a converted Humvee) traveled the
farthest distance
•But hardest terrain was at the beginning of the
course
2005 Grand Challenge:
• 132 mile race
• Passed through three narrow tunnels and
negotiated more than 100 sharp left and right
turns.
• Stanford 1st, CMU 2nd, both finished in about 6
hours
2007 Urban Grand Challenge
• Victorville, California
• Involved a 96 km urban area course
an international robotics competition
founded in 1997
RoboCupSoccer
•Simulation League
•Small-Size League
•Middle-Size League
•Standard Platform League
•Humanoid League
First Mission
• move a metallic disc from one side of an arena to
another with a completely autonomous flying robot
Second Mission
• Search for a toxic waste dump
• Map the location of partially buried randomly
oriented toxic waste drums
• Identify the contents of each drum from the hazard
labels found somewhere on the outside of each drum
• Bring a sample back from one of the drums
Third Mission
•take off
•fly to a disaster area
•search for survivors and the dead
•raging fires, broken water mains, clouds of
toxic gas, and rubble
Annual competition in artificial intelligence
that awards prizes to the chatterbot
Format of the competition is that of a
standard Turing test
Contests held on 2006 – 2011
How complicated is our brain?
•A neuron, or nerve cell, is the basic
information processing unit
•Estimated to be on the order of 10 12
neurons in a human brain
•Many more synapses (10 14) connecting
these neurons
•Cycle time: 10 -3 seconds (1 millisecond)
How complex can we make computers?
•108 or more transistors per CPU
•Supercomputer: hundreds of CPUs, 1012
bits of RAM
•Cycle times: order of 10 - 9 seconds
Conclusion
•YES: in the near future we can have
computers with as many basic processing
elements as our brain, but with
•Far fewer interconnections (wires or
synapses) than the brain
•Much faster updates than the brain
•But building hardware is very different from
making a computer behave like a brain!
 The 1996 match
Game # White Black Result
1 Deep Blue Kasparov 1–0
2 Kasparov Deep Blue 1–0

3 Deep Blue Kasparov ½–½

4 Kasparov Deep Blue ½–½

5 Deep Blue Kasparov 0–1

6 Kasparov Deep Blue 1–0

Result: Kasparov–Deep Blue: 4–2
 The 1997 rematch
Game # White Black Result
1 Kasparov Deep Blue 1–0
Deep
2 Kasparov 1–0
Blue

3 Kasparov Deep Blue ½–½

Deep
4 Kasparov ½–½
Blue

5 Kasparov Deep Blue ½–½

Deep
6 Kasparov 1–0
Blue
Result: Deep Blue–Kasparov: 3½–2½
Understanding is different to recognition
• Computers can only “see” certain types of
objects under limited circumstances
• certain constrained problems (e.g., face
recognition)
Understanding is different to recognition:
“Time flies like an arrow”
• assume the computer can recognize all the words
• how many different interpretations are there?
• time passes quickly like an arrow?
• command: time the flies the way an arrow times the
flies
• command: only time those flies which are like an
arrow
• “time-flies” are fond of arrows
• only first makes any sense
But how could a computer figure this out?
Clearly humans use a lot of implicit
commonsense knowledge in
communication
So much of what we say is beyond the
capabilities of a computer to understand at
present
• Translate text to phonetic form
• Use pronunciation rules to map phonemes
to actual sound
•sounds are not independent
•e.g., “act” and “action”
• A harder problem is emphasis, emotion
etc.
Speech synthesis, recognition and understanding
• very useful for limited vocabulary applications
• unconstrained speech understanding is still too hard

Computer vision
• works for constrained problems (hand-written zip-
codes)
• understanding real-world, natural scenes is still too
hard

Learning
• adaptive systems are used in many applications: have
their limits
Planning and Reasoning
• only works for constrained problems: e.g., chess
• real-world is too complex for general systems
Overall:
• many components of intelligent systems are
“doable”
• there are many interesting research problems
remaining
Post Office
• automatic address recognition and sorting of mail

Banks
• automatic check readers, signature verification
systems
• automated loan application classification

Customer Service
• automatic voice recognition
The Web
• Identifying your age, gender, location, from your Web
surfing
• Automated fraud detection

Digital Cameras
• Automated face detection and focusing

Computer Games
• Intelligent characters/agents
Artificial Intelligence A modern Approaach
• Second Edition (Russell & Norvig)
• Section 1.1

• Courtesy: Few slides from Md. Monirul Islam CSE,

BUET and others
Do you have
any questions ?