1. Introduction to Introduction to Artificial Intelligence Henry Kautz

2. Introductions Henry Kautz –Office hours: Tuesday 1:30-2:30, Thursday 10:30-11:30, Room 666 Daniel Lowd –Office hours: Wednesday 3:00-4:00, Room 430 Text: Russell & Norvig, AI: A Modern Approach, 2 nd Edition Sign up for class email list

3. Coursework Weekly written and short programming assignments –Posted on web page only –Read assigned chapters before class on that topic! Significant final project & write up No exams Collaboration policy: –Good to talk to other students about assignments –Write up your own solution afterward –Cite other sources of information students, web, papers

4. Discrete Inference State space search Heuristics Propositional logic Local search Constraint satisfaction Compiling to SAT First-order logic Logic programming

5. Probabilistic Inference Probability theory Bayesian networks Undirected graphical models Dynamic probabilistic models Decision theory Markov decision processes

6. Learning Decision tree learning Ensemble learning Learning graphical models Neural networks Support vector machines

7. Today History of AI State space search –DFS, BFS, IDFS –Best first –A* STRIPS planning by state space search –Assignment

8. Forerunners of AI Logic: rules of rational thought –Aristotle (384-322 BC) – syllogisms –Boole (1815-1864) – propositional logic –Frege (1848-1925) – first-order logic –Hilbert (1962-1943) – “Hilbert’s Program” –Gödel (1906-1978) – incompleteness –Turing (1912-1954) – computability, Turing test –Cook (1971) – NP completeness

9. Forerunners of AI Probability & Game Theory –Cardoano (1501-1576) – probabilities –Bernoulli (1654-1705) – random variables –Bayes (1702-1761) – belief update –von Neumann (1944) – game theory –Richard Bellman (1957) – MDP

10. Early AI Neural networks –McCulloch & Pitts (1943) –Rosenblatt (1962) – perceptron learning Symbolic processing –Dartmouth conference (1956) –Newell & Simon – logic theorist –John McCarthy – symbolic knowledge representation –Samuel's Checkers Program

11. Battle for the Soul of AI Minsky & Papert (1969) – Perceptrons –Single-layer networks cannot learn XOR –Argued against neural nets in general Backpropagation –Invented in 1969 and again in 1974 –Hardware too slow, until rediscovered in 1985 Research funding for neural nets disappears Rise of rule-based expert systems

12. Knowledge is Power Expert systems (1969-1980) –Dendral – molecular chemistry –Mycin – infectious disease –R1 – computer configuration AI Boom (1975-1985) –LISP machines –Japan’s 5 th Generation Project

13. AI Winter Expert systems oversold –Fragile –Hard to build, maintain AI Winter (1985-1990) Science went on... looking for –Principles for robust reasoning –Principles for learning

14. Symbols + Numbers Graphical probabilistic models –Pearl (1988) – Bayesian networks Machine learning –Quinlan (1975) – ID3 (aka C4.5) –Vapnik (1992) – Support vector machines –Schapire (1996) – Boosting Hot topic: statistical relational learning

15. Success Stories Deep Space One (1996) Deep Blue (1997) Countless AI systems in day to day use –Computational biology –Market research –Planning & scheduling –Hardware verification –Threat assessment

16. State-Space Search

17. Non-Optimality of Best-First Search 52 nd St 51 st St 50 th St 10 th Ave 9 th Ave 8 th Ave 7 th Ave6 th Ave5 th Ave4 th Ave 3 rd Ave 2 nd Ave SG 53 nd St Path found by Best-first Shortest Path

18. Maze Runner

19. Assignment