Copy Link
Add to Bookmark
Report
Machine Learning List Vol. 4 No. 05
Machine Learning List: Vol. 4 No. 5
Wednesday Feb. 26, 1992
Contents:
Biases in Inductive Learning
The relationship between human and machine learning
IEEE KDE: Learning and Discovery in Knowledge-Based Database
The Machine Learning List is moderated. Contributions should be relevant to
the scientific study of machine learning. Mail contributions to ml@ics.uci.edu.
Mail requests to be added or deleted to ml-request@ics.uci.edu. Back issues
may be FTP'd from ics.uci.edu in pub/ml-list/V<X>/<N> or N.Z where X and N are
the volume and number of the issue; ID: anonymous PASSWORD: <your mail address>
------------------------------
Date: Fri, 21 Feb 92 12:40:30 EST
From: gordon@aic.nrl.navy.MIL
Subject: Biases in Inductive Learning
I previously announced a workshop on "Biases in Inductive Learning"
to be held July 4 after ML-92 in Aberdeen, Scotland. Please note
that this is the correct title, rather than "... Incremental Learning"
as stated on some of the other correspondence.
I would also like to make a change to my previous announcement. We WILL
accept email submissions, but they must be in PostScript. Furthermore,
an email address for at least one author must be included in all
(email and hardcopy) submissions.
Finally, we will accept late (after March 12) submissions IF they were
not accepted by ML-92. Authors who wish to submit papers returned
from ML-92 need to send me email by March 12 so I can notify them of
their delayed submission deadline. Furthermore, authors should note
the appropriateness of their papers for a workshop before submitting
to us.
------------------------------
From: Pavan Sikka <pavan@cs.ualberta.ca>
Subject: The relationship between human and machine learning
Date: Mon, 24 Feb 1992 13:23:06 -0700
I have been working on the relationship between a number of experiments
on concept learning from the psychology literature and some existing machine
learning systems that tackle similar tasks.
I would appreciate any information on "incremental" versions of INDUCE/AQ
programs (or other similar systems). I would also appreciate pointers to any
existing work of similar nature.
Thanks.
-Pavan
email: pavan@cs.ualberta.ca
Pavan Sikka
Department of Computing Science
University of Alberta
Edmonton, Alberta
CANADA
T6G 2H1
[I'll start, but I hope others can also help.
*The most relevant paper I can think of is:
Medin, D., Wattenwaker, W., & MIchalski, R. (1987). Constraints and
preferences in inductive learning: An experimental study of human and
machine performance. Cognitive Science, 11, 299-339.
*Some other papers that compare human and machine learning methods are
Pazzani, M., & Dyer, M. (1987). A comparison of concept
identification in human learning and network learning with the
generalized delta rule. Proceedings of the Tenth International Joint
Conference on Artificial Intelligence (pp.147-151). Milan, Italy:
Morgan Kaufmann.
Pazzani, M. (1991). The influence of prior knowledge on concept
acquisition: Experimental and computational results. Journal of
Experimental Psychology: Learning, Memory & Cognition, 17, 3, 416-432.
*Some psychologists who use processes related to ML to model human learning:
Gluck, M. & Bower, G. (1988). From conditioning to category learning:
An adaptive network model. Journal of Experimental Psychology:
General, 117, 227-247.
Kruschke, J. ALCOVE: An exemplare based connectionst model of category
learning. To appear (probably has already), Psychological Review.
*And finally, some psychological work that predates most of ML, but is
realted anyway:
Bower, G., & Trabasso, T. (1968). Attention in learning: Theory and
research. New York: John Wiley.
Bruner, J. S., Goodnow, J. J., & Austin, G. A. (1956). A study of
thinking. New York: John Wiley.
*Don't forget, there's a call for papers of a special issue of the
Machine Learning Journal on Computational Models of Human Learning.
The due date is April 1.
Mike]
------------------------------
From: pattabhi@selkirk.sfu.ca (Pat Pattabhiraman)
Subject: CfPapers:IEEE TDKE Spcl Issue on Learning&Discovery in K-B DBs
Date: 24 Feb 92 02:10:46 GMT
Call For Papers
IEEE Transactions on Knowledge and Data Engineering
Special Issue on Learning and Discovery in Knowledge-Based Databases
The effective and efficient use of intelligent information systems
requires far better tools and techniques which can assist a wide range of
users to create, comprehend, modify, and otherwise use such systems.
Full use of future intelligent information systems will be impossible
without such aids. Inextricably intertwined with the rapid growth of
data and information available, techniques which extract knowledge from
databases must be developed. Techniques developed from machine learning theory
are not readily amenable to database technology without modification;
databases are rapidly changing in response to new opportunities-
hence the development of knowledge-based and object oriented paradigms.
As one noted researcher has put it, "Computers have promised us
a fountain of wisdom, but they have delivered a flood of information."
Recent research progress, coupled with reported application successes,
has created a significant interest in learning and discovery in
Knowledge-Based Databases.
The guest editors solicit contributions for this Special Issue
of TKDE in the following areas:
* Learning and Discovery in Databases
* Integration of Knowledge-based and Object-Oriented Approaches
* Data Engineering Tools and Techniques for Learning and
Discovery in Databases
* Visual and Perceptual ways of Discovery in Data
* Integration of Knowledge-based and Statistical Approaches
* Integration of Different Discovery and/or Learning Methods
* Use of Domain Knowledge in Discovery and Re-use of Discovered
Knowledge
* Learning and Discovery of Causal Structure in Data
* Interactive Systems for Data Exploration and Discovery
* High-level Query Answering and Data Summarization
* Discovery in Complex Data or Text
* Ethics of Discovery in Social Databases
* Successful Applications in Medicine, Business and other areas.
Manuscripts should be no more than 25 typewritten, double spaced
pages, including figures and references. Each manuscript should
have a title page with the title of the paper, full name(s) and
affiliation(s) of author(s), complete postal and electronic
addresses, telephone number(s), and informative 150-200 word
abstract and a list of identifying keywords.
Please submit 6 copies of a paper to the guest editors by 15 June 1992.
Nick Cercone/Mas Tsuchiya
Special Issue of IEEE TDKE
Centre for Systems Science
Simon Fraser University
Burnaby, British Columbia
Canada V5A 1S6
Acceptance status letters will be sent by 1 November 1992.
------------------------------
END of ML-LIST 4.5
