David Dowe
- School of Computer Science
& SE,
Clayton School of I.T.,
Bldg 63,
Monash University,
Clayton,
Vic. 3800, Australia
Moving office: My office re-located in the not too distant past to the
STRIP Bldg.
(the ugly grey and green thing in the middle of the car-park),
Monash University, Clayton.
But more recently I have had the privilege of being re-located again,
this time to Bldg 63.
Postdoc available (Postdoctoral Fellowship job available, deadline: 31 July 2016) :
Research
Fellow in Statistics, Machine Learning, Mixture Modelling, Latent Factor Analysis and Astrophysics
(deadline 31/July/2016)
Honours projects:
D. Dowe
2012 Honours research projects.
Ray Solomonoff (1926-2009)
85th memorial
conference (Wedn 30 Nov - Fri 2 Dec 2011),
3rd
Call for Papers,
2nd
Call for Papers,
1st
Call for Papers.
I
worked
primarily with
Prof.
Chris Wallace
- see, e.g.,
"Foreword
re C. S. Wallace",
Computer Journal,
Vol. 51, No. 5
(Sept. 2008)
[Christopher
Stewart WALLACE (1933-2004) memorial special issue],
pp523-560
(and
here).
In Chris Wallace (1933-2004)'s
posthumous
``Statistical and Inductive
Inference by Minimum Message Length''
(2005),
(a) I am given special mention in the
preface
on
page vi,
(b) I am the only living person mentioned in the
table of
contents
(also here:
pp ix-xv,
p ix,
pp x-xv),
where my name appears twice,
(c) I am the living person whose name and work are most mentioned in the
index
(also
here),
(d) other than Chris Wallace himself, (in the
reference list
[and
here])
I am the most cited author
(Wallace is cited 13 times, I am cited 7 times, nextmost is 4).
Publications:
most of my work and
publications
are in the theory and applications of the (information-theoretic)
Minimum Message Length
(MML)
principle of
statistical
and
inductive
inference
and machine learning
(and econometrics and
"knowledge discovery" and
"data mining"),
dating back to
Wallace
and
Boulton (1968).
MML and
closely related methods
(see also
Minimum Description Length
[MDL])
have much to say about the
philosophy
of science,
philosophy
of inference,
the
"Turing test",
philosophy of mind
and
intelligence;
and are useful in
medicine and many other fields.
I was Program Chair of the Information, Statistics and Induction in Science (ISIS) conference, held in Melbourne, Australia on 20-23 August 1996;
attended by
R. J. Solomonoff
(see also
obituary:
online
and
scanned),
C. S. Wallace,
J. J. Rissanen
and others.
(MML relates to the work of these people
and
Greg Chaitin,
to the field variously known as
Kolmogorov
complexity
or
algorithmic
information theory
and to
Claude
Shannon (1916-2001)'s
work on information theory
- see. e.g.,
"Minimum
Message Length and Kolmogorov complexity".)
I was an invited speaker at the iAstro
Workshop and MC Meeting
(Compliance
and Conformity, Exception and Anomaly, in Science and Engineering),
London, England, U.K., 8-9 July 2005.
Chris Wallace
and
I
are authors of the
Snob program
for unsupervised
clustering
and mixture modelling, whose documentation
you are invited to look at.
Snob does
Minimum Message Length
(MML)
mixture modelling
of Gaussian,
discrete multi-state (Bernoulli or categorical), Poisson and von Mises circular distributions.
Further details on Snob are given here.
The Snob
software
is available
- subject to conditions -
for private, academic use.
One of many other areas I have published in includes using
MML to make generalised Bayesian networks
(or generalised MML Bayes nets, or
generalised MML Bayesian nets,
or generalised MML Bayesian networks)
(or generalised directed graphical models, or
generalised MML directed graphical models)
with
a mix of both continuous and discrete variables
(Comley and Dowe, 2003)
(Comley and Dowe, 2005).
(Please see my
publications
to find out other MML-related things
that I get up to.)
I am seminar co-ordinator for the departmental
Minimum Message Length
(MML)
Research group meetings.
Introductory material that I have written on
MML
includes Wallace and
Dowe (1993), "MML estimation of the von Mises concentration parameter",
TR #93/193,
Dept of Comp Sci, Monash.
Other material includes
Wallace and Dowe (1999a),
"Minimum
Message Length and Kolmogorov complexity",
Comp.
J., Vol 42, No. 4,
pp270-283
[which is the
Computer Journal's
most downloaded ``full text as .pdf'' article -
see, e.g.,
here].
See also my Monash University
CSSE
Hons. course
CSE455
Learning and Prediction II:
MML Data Mining,
on
MML.
Publications and academic interests
My publications
are listed
here (fairly up-to-date)
and here,
and a
selected
list of some of my publications since 1992 is given here
-- requestable in printed hard copy from
either [more reliable] writing a letter to my ``snail mail'' postal address
(see above)
or [perhaps less reliable] e-mail to
enquiries At cs.monash.edu.au.
And
a brief, sketchy outline of some
of my academic interests is here
(was here).
Program committees
Previous program committees I have served on are: Information, Statistics and Induction in Science (ISIS, 1996)
(as Program Chair and General Chair),
8th Australian Joint Conf. on Artif. Intell. (1995),
4th Pacific Rim Conf. on Artif. Intell. (also 9th Austr. Joint Conf. on AI) (1996),
the 1st Pacific-Asia Conf. on Knowledge Discovery and Data Mining (PAKDD'97),
IDA-97,
5th World Meeting, International Soc. for Bayesian Analysis
(ISBA-97)
and the 10th Australian Joint Conf. on Artif. Intell. (and tutes) (1997).
I also co-chaired a track on Complexity and information-theoretic approaches to biology
for the 3rd Pacific Symposium on BioComputing (PSB-3), Hawaii, Jan. 1998;
and I was on the Program Committee for
the
6th Intelligent Systems for Molecular Biology
(ISMB '98),
the
4th International Colloquium on Grammatical Inference
(ICGI-98)
and
the 7th International
Workshop on Artificial Intelligence and Statistics (Uncertainty99).
I have also co-chaired a track on
Information-theoretic approaches to biology
(with
review form)
with
Dr. Klaus Prank
for
the
4th Pacific Symposium on BioComputing
(PSB99),
Hawaii, Jan. 1999
(with
my tutorial
and panel,
other tutorials
and
accepted papers);
and I was on the Program Committee for the
3rd Pacific-Asia
Conference on Knowledge Discovery and Data Mining
(PAKDD-99)
and
PKDD'99
(3rd Principles
and Practice of Knowledge Discovery in Databases).
Other Program Committees I have been on include:
8th International
Workshop on Artificial Intelligence and Statistics,
the
International Symposium on Adaptive
Systems (La Habana, Cuba),
the
5th Pacific-Asia Conference on
Knowledge Discovery and Data Mining
(PAKDD-01),
ICML 2002
(as area chair,
and
my ICML 2002 tutorial notes),
PRICAI-02
and
ICML 2003.
Visitors :
Dr Murray Jorgensen,
University of Waikato,
Hamilton, New Zealand; Jan - Mar 1999, Jun 2000, Nov 2000, etc.
Prof. Geoff McLachlan,
Univ. of Qld., Nov 2000.
MML tutorials and finding out more about MML
in Brisbane, Australia on 16th August '98.
Past tutorials
were given at
AI'97,
and at
ANLPF in 1998,
and an
Information theory in biology
tutorial at
PSB99;
and a
tutorial
at AI'99
on 6 December 1999 in Sydney;
and at
PRICAI'2000;
and most recently my
ICML'2002 tutorial
on 9 July 2002 at
ICML'2002.
See also
CSC423
Learning and Prediction
below.
Consulting
Consulting,
RUUG
and
RUUG Consulting.
Tertiary undergraduate (1st, 2nd and 3rd Yr), 4th Yr (Hons) and postgraduate students
First Year CS:
I was
and am now again
first year CS co-ordinator.
Fees
advice for 2001-2002 summer semester.
Third Year CS:
I am again third year co-ordinator.
Link to CSE3301: Third Year Project.
My 2006
teaching
includes
CSE455
Minimum Message Length
[which is the Comp Sci & SE
4th Year Hons. course on
on
Minimum Message Length
(MML)].
Dr David Dowe's consultation times:
Fridays at 4pm in Bldg 25 straight after the lecture.
Please contact me if interested in
my 2005 Hons. projects;
other 2005 student projects.
My postgraduate
students are :
Josh Comley,
Dean McKenzie,
both jointly or associately supervised by
Dr Lloyd Allison;
and
Greg Collie;
and
Lara Kornienko
and
Suzie Molloy,
jointly supervised with Dr David Albrecht;
and Peter Tan
and
Adrian Bickerstaffe.
Some other possible Master's and Ph.D. postgrad.
projects
are given
here,
but others are definitely possible.
My 2002 Hons. students included
Ryan Baird,
Andrew Ruan
and
Andy Serelis.
My 2001 Hons. students include
Lara Kornienko
(co-supervised with Dr David Albrecht),
Peter Jing Tan
and
Ryan Baird.
My 2000 Hons. students included
Edmund Lam
(MML),
Scott Needham
(MML
and Ockham's razor),
and
Richard Wallbrink
(chess-
and
game-player behaviour).
My 1999 Hons. students included
Lachlan Champion,
Joshua Comley,
(with Alan Dorin)
Terry Lee,
and
Cuong Ta.
My 1997 Hons. student
projects
were on :
MML factor analysis in mixture modelling (& Snob) with
R. Edwards,
(co-supervised with G. Farr) on MML inductive inference of game-player behaviour with A. Jansen, (co-supervised with G. Farr and A.J. Hurst) on
probabilistic prediction of
football with
M. Doran, and
(co-supervised with K.B. Korb, who turned up to a few meetings) on
MML Forecasting of Financial Markets with A. Pan.
Past teaching
and
past students.
Undergraduate
subject,
Honours
(project)
and
Postgraduate
Course Information.
Secondary students
I am a member of the
Australian
Informatics Olympiad committee.
Probabilistic football-tipping competition
(and
outline of
probabilistic
prediction),
free with
prizes.
Monash
Open Day
and other
School
Liaison Activities.
Other links - links to pages on:
Minimum message length
(MML)
[compare with
AIC
or with MDL],
Chris Wallace (1933-2004)
(developer of MML in
1968),
Wallace, C.S. (2005) [posthumous]
Book:
Statistical and Inductive Inference by Minimum Message
Length, Springer (Series: Information Science and Statistics), 2005, XVI,
432 pp., 22 illus., Hardcover, ISBN: 0-387-23795-X.
(Link to
table of
contents,
chapter headings and
more.)
Wallace, C.S. (with D.L. Dowe),
"Minimum
Message Length and Kolmogorov complexity",
Comp. J., Vol 42,
No. 4 (1999a),
pp270-283
[which is the
Computer Journal's
most downloaded ``full text as .pdf'' article -
see, e.g.,
here;
and which is also Chris Wallace's most cited work which is co-authored by
a still active MML researcher]
Wallace and
Dowe (1999b),
Wallace and
Dowe (1999c),
and other publications by
C. S. Wallace
and by D.L. Dowe
(on, e.g.,
MML decision trees,
MML decision graphs,
MML and
generalised hybrid Bayes nets,
MML SVM, etc.)
Bayesian belief networks,
clustering and mixture modelling,
comparisons between
MML and the subsequent MDL principle,
data repositories,
decision trees,
David Dowe publications,
Occam's razor
(Ockham's razor),
Snob
(program for MML
clustering and mixture modelling,
MML
finite mixture models),
(econometric)
time series
using MML,
medical research,
a probabilistic sports prediction
competition
(and further reading on probabilistic
scoring),
chess and game theory research;
some useful links,
TheHungerSite,
TheRainforestSite,
"do-goody"/"do-goody stuff, improving the world and saving the planet";
CSSE
Clayton
seminars;
an internal link,
FAQs
and
semester
dates;
David Dowe's computer
footy
tips
were in
Friday's The Australian,
starting Sat. 28th March 1998
through till the end of 2000.
Please e-mail me if you would like to know more.
This page,
http://www.csse.monash.edu.au/~dld,
was last updated no earlier than 23rd July 1998.
Copyright
David L. Dowe,
Monash University, Australia,
1995, 23 Apr 1998, 23 July 1998, etc.
Copying is not permitted without expressed permission from
David L. Dowe.