AKBARI@CS.COLUMBIA.EDU (John C. Akbari) (01/06/87)
ever try explaining ai concepts to people who are good programmers and
even know something about ai, but who are not real experienced in
implementing ai? they've read the intro ai books & are ready for
intermediate and advanced levels of ai wizardry.
(have encountered this several times recently, & must
confess that it's difficult, especially when you try to explain
something you've haven't hacked a lot yourself.)
so, the question is, how do you do it? in looking around, it seems that
there is not a lot of material out there between the intro ai books that
explain at high levels (& the intro lisp books that give tons of syntax)
and the papers in _artificial intelligence_. on behalf of others who
may have run into this also, i'm willing to collect suggestions.
the best sort of thing is tutorial stuff like the second half of winston
& horn's lisp book (incremental description of some of the ideas that go
into developing a simplified version of something [rule-based expert
system, atn, object-oriented system]) with enough code to play with that
actually *works*. _inside computer understanding_ is also excellent.
experimenting with the simple version seems to be very helpful in
*incrementally* understanding how to design & debug a system.
does anyone in net land have, or know of, other sources? has anyone
done this sort of thing for a course, perhaps? pointers to
tech reports, course notes, tutorials, books in progress, mini versions
of master's or dissertation work, or especially
well-documented sources for simple versions of systems that can be studied
independently (in apprenticeship mode) are all great. public domain stuff
is probably best, but licenses are ok, too. any dialect of lisp is ok, even
prolog.
topics of interest (all the usual ai stuff):
expert systems (rule-based, object-oriented, etc.)
atn's
frame systems
truth maintenance systems
machine learning
intelligent computer-assisted instruction
...
so far:
winston & horn. _lisp_ (2nd ed.). [part ii.]
charniak, riesbeck, mcdermott. _artificial intelligence programming_.
charniak & mcdermott. _intro to ai_. [sprinkled throughout]
cullingford. _natural language processing: a knowledge engineering
approach_. [lots of sources sprinkled throughout]
keravnou & johnson. _competent expert systems: a case study in fault
diagnosis_. [lots of sources at the end]
touretzky. _advanced common lisp programming_. ijcai 86 tutorial.
[higher stages of hacking karma]
schank & riesbeck. _inside computer understanding_. [mini versions of
several dissertations.]
dekleer & forbus. _truth maintenance systems_. ijcai 86 tutorial.
[tough going, no sources]
will summarize to bboard.
ad...THANKS...vance!
john c akbari
ARPANET & Internet akbari@CS.COLUMBIA.EDU
BITnet akbari%CS.COLUMBIA.EDU@WISCVM.WISC.EDU
uucp & usenet ...!seismo!columbia!cs!akbari
DECnet akbari@cs
PaperNet 380 riverside drive, no. 7d
new york, new york 10025
SoundNet 212.662.2476
[The new AI Expert magazine seems to be what you want. -- KIL]