Copy Link
Add to Bookmark
Report
AIList Digest Volume 6 Issue 090
AIList Digest Thursday, 5 May 1988 Volume 6 : Issue 90
Today's Topics:
Queries - Robustly Implemented Parser & Boyer and Moore's Prover &
Connectionist Sidewalk Rover & Causal Modeling &
Unification Recipes & KB Update Info,
Opinion - Exciting work in AI & Decision Theory in AI,
AI Tools - Training Wheels,
Academia - Leverhulem Research Fellowship
----------------------------------------------------------------------
Date: Sun, 1 May 88 22:08 EDT
From: LEWIS%cs.umass.edu@RELAY.CS.NET
Subject: robustly implemented parser wanted
This is a basic, run-of-the-mill, "We want tools" message.
In particular, what we're looking for is a robust implementation
of a syntactic parser. We're much more interested in a reliable
piece of Common LISP software that implements a basic chart parser
or something similar, than in a cutting edge piece of research
software. Plusses would be large existing grammars or lexicons,
hooks for semantic interpretation, or good documentation. The
parser would be used for research on document retrieval by
the Intelligent Information Retrieval group here at U Mass, and we
would be happy to provide the authors with plenty of data on
its performance on large quantities of real world text, as well
as on our experiences with using the on-line Longman dictionary
with it. I'd be interested in hearing from both users and
authors/maintainers of such software. If there's sufficient
interest, I'll summarize to the net.
Thanks,
David D. Lewis INTERNET: lewis@cs.umass.edu
COINS Dept. BITNET: lewis@umass
University of Massachusetts, Amherst
Amherst, MA 01003
ph. 413-545-0728
------------------------------
Date: 2 May 88 20:19:09 GMT
From: unido!laura!atoenne@uunet.UU.NET (Andreas Toenne)
Reply-to: unido!atoenne@uunet.UU.NET (Andreas Toenne)
Subject: Re: Availability of Boyer and Moore's Prover
In article <8804230021.AA05197@CLI.COM> boyer@CLI.COM (Robert S. Boyer) writes:
>A Common Lisp version of our theorem-prover is now available under the
>usual conditions: no license, no copyright, no fee, no support. The
>To get a copy follow these instructions:
>
>1. ftp to Arpanet/Internet host cli.com.
Ouch!
Can anyone send me this theorem-prover by e-mail please?
I have no ftp access :-(
Please send it to the following BITNET Address, as UUCP (and Arpa) Mail
costs me real $$.
atoenne@ddoinf6.bitnet
Thank you in advance
Andreas Toenne
------------------------------
Date: 3 May 88 18:48:55 GMT
From: gary%desi@ucsd.edu (Gary Cottrell)
Reply-to: desi!gary@ucsd.edu (Gary Cottrell)
Subject: Re: Need info on new CMU sidewalk rover
In article <8804301216.AA23619@ucbvax.Berkeley.EDU> jbn@GLACIER.STANFORD.EDU
(John B. Nagle) writes:
>
> I hear that CMU has a new autonomous vehicle, a sidewalk rover
>like the Terregator, but improved. Who is doing the work, and are there
>any papers yet?
>
> John Nagle
I don't know who is working on it, but I heard from a usually reliable
source that Dean Pomerleau (grad student at CMU, inventor of meta-connection
networks) has a 3-layer back-prop network driving the thing better than the
CMU vision group's system. Anyone care to confirm or deny this information?
gary cottrell
Computer Science and Engineering C-014
UCSD,
La Jolla, Ca. 92093
gary@sdcsvax.ucsd.edu (ARPA)
{ucbvax,decvax,akgua,dcdwest}!sdcsvax!sdcsvax!gary (USENET)
gwcottrell@ucsd.edu (BITNET)
------------------------------
Date: Tue 3 May 88 17:44:47-PDT
From: SCHWARTZ@PLUTO.ARC.NASA.GOV
Subject: Causal modeling
From: PLU::SCHWARTZ 29-APR-1988 14:43
To: RI
Subj: Causal Modeling
For a survey white paper for HQ-OAST, I am trying to gather information on
all of the causal modeling efforts being pursued at ARC. Specifically, the
information I am seeking is:
Who is doing work which concerns causal modeling and AI?
What domain/application is being used to pursue this work?
Why are Causal Modeling techniques being used?
What is the level of effort associated with the work (manpower/resources)?
Is the work being done in-house (NASA) or on grant/contract outside of NASA?
Are there any papers ( a research proposal will suffice) that are available
concerning your work?
Please help me in collecting this information. One liners as answers are fine.
I'd li
It is imperative that I have this information by May 6th. Thanks for your
support!
Mary Rudokas
------------------------------
Date: Mon, 2 May 88 13:06:17 EDT
From: spike%mwcamis@mitre.arpa
Subject: unification recipes wanted
--------
If you happen to have a reference list handy on the topic of
unification algorithms in pattern matching systems, I'd be
grateful to obtain such.
[On a different note, I send my gratitude to our AIList moderator
Kenneth Laws for your impressive and sustained effort in providing
an incredibly useful service.]
Thanks, Ben Bloom (spike@mitre.arpa)
------------------------------
Date: 3 May 88 10:45:00 CST
From: "HENRY::TSATSOUL" <tsatsoul%henry.decnet@space-tech.arpa>
Reply-to: "HENRY::TSATSOUL" <tsatsoul%henry.decnet@space-tech.arpa>
Subject: KB Update Info Request
I am interested in investigating the updating of KBs. In other words,
when a system makes inferences to solve a goal or answer a question, it
generates many intermediate facts. The question is which of these to keep,
how, when, where, for how long, etc. Do you know of any work related to
this? (I am not interested in truth maintenace!!)
Thank you,
--------------------------------------------------------------------------------
| Costas Tsatsoulis | tsatsoul @ space-tech.arpa |
| Dept. of Electrical and Computer Engineering |-------------------------------|
| Nichols Hall | |
| The University of Kansas | H A I K U |
| Lawrence, KS 66045 | |
------------------------------
Date: 2 May 88 20:42:46 GMT
From: stuart%warhol@ads.com (Stuart Crawford)
Reply-to: stuart@ads.com (Stuart Crawford)
Subject: Re: Exciting work in AI
Wray Buntine (wray@nswitgould.oz) writes:
> Ross's original ID3 work (and the stuff usually reported in Machine Learning
> overviews) and much subsequent work by him and others (e.g. pruning)
> actually fails the "real AI" test. It was independently developed by
> a group of applied statisticians in the 70's and is well known
> Breiman, L., Friedman, J.H., Olshen, R.A. and Stone, C,J. (1984)
> "Classification and Regression Trees", Wadsworth
> Ross's more recent work does significantly improve on Breiman et al.s stuff.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
How? If you mean his stuff on generating production rules from decision trees,
I think you're missing the point of CART. It seems to me that simply
transforming decision trees into production rules is a rather uninteresting
exercise. Quinlan tries to motivate the idea by suggesting that the generated
rules are an "improvement" over the induced tree because they are both easier
to interpret and more parsimonious. I disagree that they are easier to
interpret, and they are more parsimonious only if your original induction
algorithm has not already pruned the tree. Using production rule generation as
an alternative to tree pruning strikes me as the wrong approach. I still feel
that CART is the induction procedure of choice because of the following:
1. generates parsimonious trees
2. handles noisy, incomplete data
3. strong, well understood, asymptotic properties
4. allows user-defined priors and cost-functions
5. delivers attribute-importance diagnostics
6. can induce rules incrementally
7. delivers low bias, low variance estimates of misclassification rate
For references on 1-5, see Brieman et al. (1984), and for 6,7 see Crawford, S.
"Extensions to the CART Algorithm", proceedings Knowledge Acquisition for
Knowledge-Based Systems workshop (1987).
I also find somewhat curious Buntine's suggestion that Quinlan's most recent
work,
> is closer to real AI (e.g. concern for comprehensibility),
> though it still has an applied statistics flavour.
I would suggest that the work has an applied statistics flavor because it is
attempting to solve an applied statistics problem.
--------------------------------------
Stuart Crawford
stuart@ads.com
Advanced Decision Systems
1500 Plymouth Street
Mountain View, CA 94043
Stuart
------------------------------
Date: 2 May 88 21:27:51 GMT
From: mikeb@wdl1.UUCP (Michael H. Bender)
Subject: Re: Decision Theory in AI.
You might want to take a look at Judea Pearl's (UCLA) work on the use
of Influence Diagrams for uncertainty reasoning. It is closely
associated with work by Ron Howard (Stanford) and others.
Mike Bender (mikeb@ford-wdl1)
------------------------------
Date: 3 May 88 01:37:30 PDT (Tuesday)
From: "Elizabeth_Lloyd.WGCERX"@Xerox.COM
Subject: Training wheels
I do have some firm references for the above work, but they're not with me now.
However a few pointers - your time frame is correct - I think you should search
the following journals - HUMAN FACTORS; IBM SYSTEMS JOURNAL; BEHAVIOUR AND
INFORMATION TECHNOLOGY; INTERNATIONAL JOURNAL OF MAN-MACHINE STUDIES. I can't
remember which of these contains the information you want, but if you don't have
any sucess, then contact me directly and I'll search out and send you the exact
references.
Elizabeth
------------------------------
Date: 3-MAY-1988 13:54:01 GMT
From: HANCOXPJ%ASTON.AC.UK@CUNYVM.CUNY.EDU
Subject: Leverhulem research fellowship
From: Dr P J Hancox <HANCOXPJ@uk.ac.aston>
Dept: Computer Science
Tel No: 021 359 3611 X4652
Leverhulme Fellowship
The Computer Science Department of Aston University would welcome enquiries
concerning a Leverhulme Fellowship (about 1 year) from active researchers
who has been awarded a PhD in the last five years. The other restriction is
that the Fellow must be from the US or the UK Commonwealth.
Aston University is a technological institution in the centre of Birmingham
(UK). As an institution, it is dedicated to the application of information
technology. The Department has strong research interests in NLP (esp LFG),
image processing, information systems engineering and software engineering.
Those interested should initially contact Peter Hancox
(hancoxpj@mail.aston.ac.uk / hancoxpj@aston.uucp). All messages will be
acknowledged.
Peter Hancox
------------------------------
End of AIList Digest
********************
