1. Course Overview  What is AI?  What are the Major Challenges?  What are the Main Techniques?  Where are we failing, and why?  Step back and look at the Science  Step back and look at the History of AI  What are the Major Schools of Thought?  What of the Future? Done

2. Course Overview  What is AI?  What are the Major Challenges?  What are the Main Techniques? (How do we do it?)  Where are we failing, and why?  Step back and look at the Science  Step back and look at the History of AI  What are the Major Schools of Thought?  What of the Future?

3. Course Overview  What is AI?  What are the Major Challenges?  What are the Main Techniques? (How do we do it?)  Where are we failing, and why?  Step back and look at the Science  Step back and look at the History of AI  What are the Major Schools of Thought?  What of the Future?  Search  Logics (knowledge representation and reasoning)  Planning  Bayesian belief networks  Neural networks  Evolutionary computation  Reinforcement learning

4. Course Overview  What is AI?  What are the Major Challenges?  What are the Main Techniques? (How do we do it?)  Where are we failing, and why?  Step back and look at the Science  Step back and look at the History of AI  What are the Major Schools of Thought?  What of the Future?  Search  Logics (knowledge representation and reasoning)  Planning  Bayesian belief networks  Neural networks  Evolutionary computation  Reinforcement learning

5. Course Overview  What is AI?  What are the Major Challenges?  What are the Main Techniques? (How do we do it?)  Where are we failing, and why?  Step back and look at the Science  Step back and look at the History of AI  What are the Major Schools of Thought?  What of the Future?  Search  Logics (knowledge representation and reasoning)  Planning  Bayesian belief networks  Neural networks  Evolutionary computation  Reinforcement learning These are all in fact types of “Machine Learning”

6. Dealing with Uncertainty  The need to deal with uncertainty arose in “expert systems”  Code expertise into a computer system Example:  Medical diagnosis: MYCIN  Equipment failure diagnosis in a factory  Sample from MYCIN:  IF  The infection is primary-bacteremia AND  The site of the culture is one of the sterile sites AND  The suspected portal of entry is the gastrointestinal tract  THEN  There is suggestive evidence (70%) that the infection is bacteroid  Expert systems often have long chains  IF X THEN Y … IF Y THEN Z … IF Z THEN W …  If uncertainty is not handled correctly, errors build up, wrong diagnosis  Also, there may be dependencies, e.g. X and Y depend on each other  Leads to more errors…  Need a proper way to deal with uncertainty

7. How do Humans Deal with Uncertainty?  Not very well…  Consider a form of cancer which afflicts 0.8% of people (rare)  A lab has a test to detect the cancer  The test has a 98% chance to give an accurate result  Mr. Bloggs goes for the test  The result comes back positive  i.e. the test says he has cancer  What is the chance that he has the cancer?  28%  Afflicts experts too  Studies have shown: human experts thinking of likelihoods do not reason like mathematical probability

8. A BC DE Increased total serum count Metastatic cancer Brain Tumour Severe headaches Coma

9. A BC DE Increased total serum count Metastatic cancer Brain Tumour Severe headaches Coma No Link

10. A BC DE Increased total serum count Metastatic cancer Brain Tumour Severe headaches Coma Serum count Brain tumour Coma Yes 95% YesNo94% NoYes29% No 0.1%

11. A BC DE Increased total serum count Metastatic cancer Brain Tumour Severe headaches Coma Serum count Brain tumour Coma Yes 95% YesNo94% NoYes29% No 0.1% Brain tumour headache Yes70% No1%

12. A BC DE Increased total serum count Metastatic cancer Brain Tumour Severe headaches Coma Serum count Brain tumour Coma Yes 95% YesNo94% NoYes29% No 0.1% Brain tumour headache Yes70% No1% …… …… …… …… …… …… …… …… ……

13. Inference in Belief Networks  Questions for a belief network:  Diagnosis  Work backwards from some evidence to a hypothesis  Causality  Work forwards from some hypothesis to likely evidence  Test a hypothesis, find likely symptoms  In general – mixed mode  Give values for some evidence variables  Ask about values of others  No other approach handles all these modes  Reasoning can take some time  Need to be careful to design network  Local structure: few connections

14. How Good are Belief Networks?  Relieves you from coding all possible dependencies  How many possibilities if full network?  Tools are available  Build network graphically  System handles mathematical probabilities  Case study:  Pathfinder a medical expert system  Assists pathologists with diagnosis of lymph-node diseases  Pathfinder is a pun  User enters initial findings  Pathfinder lists possible diseases  User can  Enter more findings  Ask pathfinder which findings would narrow possibilities  Pathfinder refines the diagnosis  Pathfinder version based on Belief Networks performs significantly better than human pathologists

15. QUIZ

16. “Machines will be capable, within _____ years, of doing any work that a man can do.” Herbert Simon, 1965.

17. What disciplines are considered as sub-areas of Cognitive Science?

18. 1966: “Any task that requires real understanding of natural language is too difficult for a computer” - Bar-Hillel

19. Time flies like an arrow.

20. Give an example which shows why speech recognition is hard for computers.

21. Major tasks in robotics: 1.Localisation/mapping  Range finders  Landmarks  Always uncertainty 2.?

22. DARPA Grand Challenge 2004 150-mile route in Mojave Desert (off-road course) Best performance? (require accuracy to nearest 2 miles)

23. Vision Hierarchy 4. High level Models 3. Mid level ???? 2. Putting together Multiple images 1. Low level processing on a single image 0. The physics of image formation

24. Go (Wei Qi) Humans don’t want to play computers because ?

25. John McCarthy, "Programs with Common Sense", 1958. "Our ultimate objective is to make programs that learn from their experience as effectively as humans do. We shall…say that a program has common sense if ?”

26. Defence A big user of AI. "... the deployment of a single ?????? called DART during the Desert Shield/Storm Campaign paid back all US government investment in AI/KBS research over a 30 year period." Tate A. Smart Planning. ARPI Proc. 1996.

27.  Can use logic to represent a hierarchy of concepts  isa(Tweety, canary)  isa(canary, bird)  isa(bird, animal)  isa(animal, living_thing)  isa(living_thing, physical_thing)  isa(physical_thing, tangible_thing)  isa(tangible_thing, thing)  What do we call this?

28. “Machines will be capable, within _____ years, of doing any work that a man can do.” Herbert Simon, 1965.

29. “Machines will be capable, within twenty years, of doing any work that a man can do.” Herbert Simon, 1965.

30. What disciplines are considered as sub-areas of Cognitive Science?

31. Cognitive Science  Psychology  Philosophy  Neuroscience  Artificial Intelligence  Linguistics  Anthropology

32. 1966: “Any task that requires real understanding of natural language is too difficult for a computer” - Bar-Hillel

34. Time flies like an arrow.

35.  (You should) time flies as you would (time) an arrow  Time flies in the same way that an arrow would (time them)  Time those flies that are like arrows  Fruit flies like a banana  each of above  Time magazine travels straight when thrown

36. Give an example which shows why speech recognition is hard for computers.

37.  “eat I scream” vs. “eat ice cream”

38. Major tasks in robotics: 1.Localisation/mapping  Range finders  Landmarks  Always uncertainty 2.?

39. Major tasks in robotics: 1.Localisation/mapping  Range finders  Landmarks  Always uncertainty 2.Motion planning  For body location in world  For arms/fingers

40. DARPA Grand Challenge 2004 150-mile route in Mojave Desert (off-road course) Best performance? (require accuracy to nearest 2 miles)

41. DARPA Grand Challenge 2004 150-mile route in Mojave Desert (off-road course) Best performance: 7.36 miles

42. Vision Hierarchy 4. High level Models 3. Mid level ???? 2. Putting together Multiple images 1. Low level processing on a single image 0. The physics of image formation

43. Vision Hierarchy 4. High level Models 3. Mid level Segmentation 2. Putting together Multiple images 1. Low level processing on a single image 0. The physics of image formation

44. Go (Wei Qi) Humans don’t want to play computers because ?

45. Go (Wei Qi) Humans don’t want to play computers because computers are too bad

46. John McCarthy, "Programs with Common Sense", 1958. "Our ultimate objective is to make programs that learn from their experience as effectively as humans do. We shall…say that a program has common sense if ?”

47. John McCarthy, "Programs with Common Sense", 1958. "Our ultimate objective is to make programs that learn from their experience as effectively as humans do. We shall…say that a program has common sense if it automatically deduces for itself a sufficient wide class of immediate consequences of anything it is told and what it already knows.”

48. Defence A big user of AI. "... the deployment of a single ?????? called DART during the Desert Shield/Storm Campaign paid back all US government investment in AI/KBS research over a 30 year period." Tate A. Smart Planning. ARPI Proc. 1996.

49. Defence A big user of AI. "... the deployment of a single logistics support aid called DART during the Desert Shield/Storm Campaign paid back all US government investment in AI/KBS research over a 30 year period." Tate A. Smart Planning. ARPI Proc. 1996.

50.  Can use logic to represent a hierarchy of concepts  isa(Tweety, canary)  isa(canary, bird)  isa(bird, animal)  isa(animal, living_thing)  isa(living_thing, physical_thing)  isa(physical_thing, tangible_thing)  isa(tangible_thing, thing)  What do we call this?

51.  Can use logic to represent a hierarchy of concepts  isa(Tweety, canary)  isa(canary, bird)  isa(bird, animal)  isa(animal, living_thing)  isa(living_thing, physical_thing)  isa(physical_thing, tangible_thing)  isa(tangible_thing, thing)  What do we call this?  Ontology