Lecture 1 on Artificial Intelligence as taught by Harsh Pandya

1. Duke TIP (GaTech Campus)
Instructor: Harsh Pandya
Teaching Assistant: Taylor Pospisil

2. Who is Intelligent? And
why?
• What is Intelligence?
• What is Artificial Intelligence?
• History of AI

3. Intelligence
 Knowledge
 Logical
 Reasonable
 Or just Human?

4. Artificial Intelligence
Designing
machines that:
 Think like human
 Act like human
 Think Rationally
 Act Rationally
 Being rational means
maximally achieving
your pre-defined goals

5. Why not mimic human
brain?
“Brains are to Intelligence as wings are to
flight”
Brains are good at rational thinking but not
perfect

6. The Turing Test

7. Footprints of AI
Dates back to 8th
century BC,
greek mythology.
Hephaestus
created Talos to
guard crete.

8. Footprints of AI
 Aristotle 4th century BC, developed an
informal logic system
 Descartes, 17th century AD proposed
animal bodies are complex machines
 Blaise Pascal, 1642, first mechanical
digital calculator
 19th century, George Boole developed
binary algebra better known as boolean
algebra
 Charles Babbage worked on
programmable calculators

9. Footprints of AI
 1943, McCulloch & Pitts: Boolean circuit model
of brain
 1950, Alan Turing published “Computing
Machinery and Intelligence” introduced concept
of learning in machines by letting them alter their
program
 1950s, Samuel's checkers program, Newell &
Simon's Logic Theorist, Gelernter's Geometry
Engine
 1956: Dartmouth meeting: the term “Artificial
Intelligence” adopted
 1965: Robinson's complete algorithm for logical
reasoning

10. Footprints of AI
 1980-88: Expert systems industry
booms
 1988-93: Expert systems industry
busts: “AI Winter”
 1990s: Resurgence of probability,
focus on uncertainty
 Neural Networks became more
popular, Agents and learning systems
bloomed
 1997, Deep Blue beats Gary Kasparov

11. Realms of AI
Introducing the sub-domains

12. Natural Language
processing
 Automatics Speech recognition
 iPhone Siri / OK Google!
 Text to Speech converter (TTS)
 Machine Translation (Language Transaltor)
 e.g. Google Translator
 Spam Filtering
 Web search (Google!)

13. Computer Vision
 Object and Face recognition
 Motion tracking
 Mapping unknown environment
 Extended vision – UV, IR, X-ray

14. Robotics
 Autonomous vehicles
 Humanoids
 Petman, iCUB, THOR
 Prosthetics
 Sports/Athletics
 Personal robots
 Cyborg

15. Game Playing
 Simple games
 Tic Tac Toe, Pacman
 Kasparov vs Deep blue
 Modern Computer & Console games
 IBM Watson winning Jeopardy

16. Decision making systems
 Scheduling,
 e.g. airline routing, military
 Route planning, e.g. Google maps
 Medical diagnosis
 Web search engines
 Credit card fraud detection
 Product recommendations / Pop up
ads

17. What AI can and cannot
 Play sports like soccer/Table tennis?
 Play a decent game of Jeopardy?
 Drive safely along a curving mountain road?
 Buy a week's worth of groceries on the web?
 Discover and prove a new mathematical theorem?
 Converse successfully with another person for an hour?
 Perform a surgical operation?
 Put away the dishes and fold the laundry?
 Translate spoken Chinese into spoken English in real time?
 Write an intentionally funny story?
 Understand human emotions and imitate them?

18. What AI can and cannot
 Detect cancer?
 Plan a journey to Andromeda
galaxy?
 Surf the internet?
 Create aesthetic music?
 Write a creative poetry?

19. Time-warp
 What is Intelligence?
 Artificial Intelligence & its
classification
 Turing test
 History of AI
 Subdomains of AI
 What AI can and cannot

20. Questions?