00100	
00200	
00300	
00400	
00500	
00600	
00700	
00800	
00900	                         STANFORD UNIVERSITY
01000	
01100	
01200	                     COMPUTER SCIENCE DEPARTMENT
01300	
01400	
01500	                            April 1, 1972

01600	
01700	
01800	
01900	
02000	
02100	
02200	
02300	
02400	
02500	                    Ph.D. QUALIFYING EXAMINATION

02600	
02700	
02800	                       Artificial Intelligence

02900	
03000	
03100	
03200	
03300	
03400	
03500	
03600	
03700	The examination will be open book.  The first session  will  be  from
03800	9:30  to  12:30  pm, and the second session will be from 1:30 to 4:30
03900	pm.  No work on the exam is to be done during the lunch break.
     

00100	
00200	   SYLLABUS FOR QUALIFYING EXAMINATION IN ARTIFICIAL INTELLIGENCE
00300	    Computer Science Department, Stanford University, April, 1972
00400	
00500	
00600		Ideally, a student preparing to take a qualifying examination
00700	in artificial intelligence would be advised  to  read  Grimblefritz's
00800	"Methods  and  Issues  in  Artificial  Intelligence, 5th edition" and
00900	would be assured that if he had  read  this  book  and  mastered  the
01000	techniques  described  therein, he could pass the examination even if
01100	to do research in AI would  require  acquaintance  with  more  recent
01200	work.   Unfortunately, no such book exists, partly because no-one has
01300	written it, but mainly because the field is not in a state that would
01400	permit it.
01500	
01600		Nevertheless, this year the AI Quals Committee has decided to
01700	try a short reading list using one or two references for each of  the
01800	main subfields of artificial intelligence.
01900	
02000		For  the  purposes  of  this  exam  we  will  divide  AI into
02100	subfields as follows:
02200	
02300		1. Heuristics.  This includes methods for searching spaces of
02400	possibilities  for  solutions  to problems.  It includes game playing
02500	and theorem proving as subfields.  The general reference is  (Nilsson
02600	1971)  for  general methods.  However, much of the information in the
02700	field is contained in separate  investigations  that  have  not  been
02800	generalized.   Therefore,  we  also ask you to read (Slagle 1971) and
02900	(Feigenbaum, Buchanan and Lederberg  1971).   Search  algorithms  are
03000	often  written  in  languages devised for AI purposes such as LISP or
03100	more recently Microplanner (Winograd et al, 1971b).  General  purpose
03200	languages  like  Algol  are also used.  There may be questions on the
03300	exam requiring that a program be written  expressing  some  heuristic
03400	procedure,   and   familiarity  with  LISP,  Microplanner,  and  some
03500	Algol-like language is expected.
03600	
03700		2.  Representation.   It is becoming increasingly clear  that
03800	the  representation  of information in the machine about the external
03900	world, the laws that govern the effects of actions, goals, and  where
04000	knowledge  is  to be found is a key problem in AI.  At present, there
04100	is not a uniformly used approach to this.  The student is expected to
04200	understand  the  approaches  described  in (McCarthy and Hayes 1969),
04300	(Winograd 1971a), (Newell 1965), and (Amarel 1968).
04400		A  related  area  is  the  processing of information given in
04500	natural language.   Here the references are (Winograd  1971a)  again,
04600	(Simmons
04700	
04800		3.  Robotics.  This  includes  methods of getting information
04900	about the physical world into the computer  and  representing  it  in
05000	useful  form  once  it is gotten in.   The references here are not in
05100	particularly good shape but the student should read (Newell 1971, the
05200	Speech Report), (Falk 1970), and (Feldman et. al. 1971).
05300	
05400		The  current  state  of  research in AI is represented in the
05500	journal  "Artificial  Intelligence",  the  Proceedings  of   the   AI
05600	Conferences  (1969  and  1971),  the  volumes  in  the series Machine
05700	Intelligence.  Collections of older work  are  included  in  Minsky's
05800	Semantic  Information  Processing and Computers and Thought edited by
05900	Feigenbaum and Feldman.
06000	
06100	
06200	                             REFERENCES
06300	
06400	
06500	Amarel, S. (1968) On Representations of Problems of Reasoning about
06600		Actions. Machine Intelligence 3, pp. 131-171 (eds Meltzer,
06700		B. and Michie, D.). New York: American Elsevier Publishing
06800		Company, Inc.
06900	
07000	Buchanan, B., Feigenbaum, E.A. and Lederberg, J. (1971) A Heuristic
07100		Programming Study of Theory Formation in Science. Proceedings
07200		of the Second International Joint Conference on Artificial
07300		Intelligence. London:The British Computer Society.
07400	
07500	Falk, Gilbert (1970) Computer Interpretation of Imperfect Line Data
07600		as a Three-dimensional Scene.     Stanford Artificial
07700		Intelligence Report:Project Memo AI-132.
07800	
07900	Feigenbaum, E.  and Feldman, J. (eds) (1963) Computers  and  Thought.
08000		New York:McGraw-Hill.
08100	
08200	Feldman, J., Pingle, K., Binford, T., Falk, G., Kay,, A., Paul, R.,
08300		Sproull, R. and Tenenbaum, J. (1971) The Use of Vision and
08400		Manipulation  to  Solve  the   "Instant   Insanity"   Puzzle.
08500		Proceedings of the Second International Joint Conference on
08600	        Artificial Intelligence. London:The British Computer Society.
08700	
08800	McCarthy,  John and Hayes, P. (1969) Some Philosophical Problems from
08900	        the Standpoint  of Artificial Intelligence.  Machine 
09000		Intelligence 4, pp. 463-502 (eds Meltzer, B. and Michie, D.).
09100		Edinburgh:Edinburgh University Press.
09200	
09300	Minsky,  Marvin  (ed)   (1968)   Semantic   Information   Processing.
09400		Cambridge:M.I.T. Press.
09500	
09600	Newell, A.  (1965) Limitations of the Current Stock  of  Ideas  about
09700	        Problem-Solving.     Proceeddngs  of  a  Conference   on
09800		Electronic Information Handling, pp. 195-208, (eds Kent,
09900	        A. and Taulbee,  O.).   New York: Spartan.
10000	
10100	Newell, A., Barnett, J., Forgie, J., Green,  C.,  Licklider,  J.C.R.,
10200		Munson, J., Reddy, R. and Woods, W. (1971) Speech
10300		Understanding Systems: Final Report of a Study Group.
10400	        Carnegie-Mellon  University:Computer Science Department
10500	
10600	Nilsson,  N.J.     (1971)  Problem-Solving  Methods   in   Artificial
10700	        Intelligence .New York: McGraw-Hill.
10800	
10900	Simmons, R.  (1970)  Natural  Language  Question  Answering  Systems.
11000		Communications of the ACM, 13, 1, 15-30.
11100	
11200	Slagle,  J.R.     (1971)  Artificial  Intelligence:     The Heuristic
11300		Programming 	Approach. New York:McGraw-Hill.
11400	
11500	Winograd, Terry (1971) Procedures as a Representation for Data  in  a
11600	        Computer Program for Understanding Natural Language.   Ph.D.
11700		Thesis, M.I.T.
11800	
11900	Winograd, T. and Sussman, G.J. (1971) Micro-Planner Reference Manual
12000		(available at the A.I. Project)
12100