(Chris) C. S. Wallace's Publications

Chris Wallace (26 October 1933 – 7 August 2004)

BIB(CSW): Christopher Stewart Wallace's publications data-base can be searched for various topics:

—query

search results:

output area

—op

[isbn:038723795X]

You might like to try some of these keywords, singly or in (small) combinations:: algorithm, analysis, application, architecture, Bayesian, binomial, bioinformatics, biology, book, c195x, c196x, c197x, c198x, c199x, c200x (-- decades), capability, causal, classification, cluster, COMPJ, computer, decision, distribution, entropy, estimator, factor, file, FSMML, gas, Gaussian, graph, hardware, hierarchic, inference, information, input, jrnl, JRSS, medical, methods, MML, multiplier, multiprocessor, Nature, network, normal, numerical, operating, physics, Poisson, PRNG, probability, random, RNG, SMML, Snob, system, taxonomy, theory, TR, tree, Walnut.
NB. No wildcards, "and", or "or"!
(The collection may be incomplete.)

Chris Wallace was appointed Foundation Chair of Computer Science at Monash University in 1968, and Professor Emeritus in 1996.

[Monash archives]	Les G & [CSW c1991] courtesy of James (Jimmy) Wilkinson.
With the "waving arm", [M. Memo] 13/10/2004


%T Chris Wallace (C. S.)
%D 1933-2004
%K Wallace, CSW, minimum message length, MML, computer science, Australia, MDL,
   CSWallace, CSSE, Monash, obit, description, information theory, zz1004
%X [<A HREF="/~lloyd/tildeImages/People/Wallace/">C S Wallace's publications</A>]['04].
   [<A HREF="http://www.csse.monash.edu.au/~gopal/csw.html">obit</A>]['04] by Gopal Gupta.

%A D. L. Dowe
%A L. Allison
%A T. I. Dix
%A L. Hunter
%A C. S. Wallace
%A T. Edgoose
%T Circular clustering of protein dihedral angles by minimum message length
%J Pacific Symposium on Biocomputing '96
%M JAN
%P 242-255
%D 1996
%I World Scientific
%O TR 95/237, Dept. Computer Science, Monash University, Oct 1995
%K PSB, PSB96, TR 237, TR237, Monash, DLD, CSW, CSWallace, LAllison, MolBio,
   Monash, classification, angle, von Mises, vonMises, protein structure,
   inductive inference, II, MML, MDL, conf, bioinformatics, c1996, c199x, c19xx
%X L. Hunter - NLM, NIH. PSB '96: 3-6 Jan 1996, Hawaii; uk us isbn:9810225784.
   [<A HREF="http://www.allisons.org/ll/Publications/1996PSB/">paper</A>],
   [<A HREF="http://psb.stanford.edu/psb96/proceedings/dowe.ps">paper.ps</A>][1/'96],
   [<A HREF="http://www.cgl.ucsf.edu/psb/psb96/proceedings/eproceedings.html">[eProceedings]</A>][1/'96].

%A L. Allison
%A C. S. Wallace
%T An information measure for the string to string correction problem with
   applications
%J 17th Australian Comp. Sci. Conf.
%P 659-668
%M JAN
%D 1994
%W Christchurch, N. Z.
%K LAllison, CSW, CSWallace, Monash, conf, MolBio, inductive inference, II,
   string, sequence, family, evolutionary, phylogenetic, tree, trees,
   variation, variance, uncertainty, estimate, estimation, parameters, DNA,
   multiple alignment, Gibbs sampling, sample, GS, simulated annealing SA,
   minimum message length MML, Bayesian, temperature, cooling, probabilistic,
   NZ, New Zealand, c1994, c199x, c19xx, ACSC 17, 94, ACSC17, ACSC94,
   bioinformatics, Monash
%O Australian Comp. Sci. Comm., Vol 16,  No 1(C), 1994, isbn:047302313X.
%X It has been shown how to calculate a probability for an alignment.
   Alignments are sampled from their posterior probability distribution.
   This is extended to multiple alignments (of several strings).  Averaging
   over many such alignments gives good estimates of how closely the strings
   are related and in what way.  In addition, sampling in an increasingly
   selective way gives a simulated annealing search for an optimal alignment.
   [<A HREF="http://www.allisons.org/ll/Bioinformatics/Multiple/">Bioinformatics</A>],
   [<A HREF="/~lloyd/tildeStrings/Multiple/94.ACSC-17.html">paper</A>].
   See also the related paper J. Mol. Evol. (39, pp418-430, 1994),
   "The posterior probability distribution ...", for more results.

%A L. Allison
%A C. S. Wallace
%T The posterior probability distribution of alignments and its application
   to parameter estimation of evolutionary trees and to optimization of
   multiple alignments
%J J. Mol. Evol.
%V 39
%N 4
%P 418-430
%M OCT
%D 1994
%O An earlier version is TR 93/188, Dept. Comp. Sci., Monash U., July '93
%K jrnl, MolBio, JME, c1994, c199x, c19xx, LAllison, CSWallace, CSW, DNA,
   bioinformatics, optimisation, estimate, infer, parameters, algorithm,
   multiple, alignment, data, string, molecular, sequence, homology, Markov,
   family, phylogenetic, tree, trees, edit distance, Monte Carlo method, mcmc,
   simulated annealing, SA, inductive inference, II, sample, speed, Bayesian,
   dynamic programming algorithm, DPA, stochastic, methods, GS, Gibbs sampling,
   minimum message length encoding, MML, chain, minimum description length, MDL,
   transthyretin, chloramphenicol resistance gene, CAT, CATB, CATD, CATP, CATQ,
   CCOLI, ECOLI, algorithmic, mutual information, theory, significance,
   probabilistic, temperature, limits, TR 93/188, TR188
%X <!-- ISI j's IF=1.7 at '06 --> "It is shown how to sample alignments from their posterior probability
   distribution given two strings.  This is extended to sampling alignments of
   more than two strings.  The result is firstly applied to the estimation of
   the edges of a given evolutionary tree over several strings.  Secondly,
   when used in conjunction with simulated annealing, it gives a stochastic
   search method for an optimal multiple alignment."
   -- [<A HREF="http://www.allisons.org/ll/Publications/1994JME/">paper</A>] and source code,
      [<A HREF="/~lloyd/tildeStrings/Multiple/94.JME/PP.ps">reprint.ps</A>],
      [<A HREF="https://doi.org/10.1007/BF00160274">doi:/10.1007/BF00160274</A>]['07].
   (The JME paper is a much expanded and changed version of TR 93/188,
    [<A HREF="/~lloyd/tildeStrings/Alignment/93.188.TR.ps">TR93/188</A>](.ps))

%A D. L. Dowe
%A J. Oliver
%A L. Allison
%A T. I. Dix
%A C. S. Wallace
%T Learning rules for protein secondary structure prediction
%J Proc. 1992 Department Research Conf.
%I Dept. Computer Science, University of Western Australia
%E C. McDonald
%E J. Rohl
%E R. Owens
%M JUL
%D 1992
%O TR 92/163, Dept. Computer Science, Monash University, JUN '92
%K LAllison, CSW, DLD, Monash, UWA, WA, conf, MolBio, decision tree, trees,
   graph, protein, amino acid, AA, secondary structure, SS, prediction,
   rule, rules, alpha helix, beta strand, extended sheet, coil, turn,
   CSWallace, inductive inference, II, MML, minimum message length, c1992,
   c199x, c19xx, bioinformatics, TR 92 163, TR92-163, TR163
%X [<A HREF="/~lloyd/tildeStrings/Protein/93.HICSS_26/92.163.TR.ps">TR92/163.ps</A>]
   Also see [<A HREF="http://www.allisons.org/ll/Bioinformatics/">Bioinformatics</A>],
   and TR 92/163.
   [<A HREF="http://www.cs.uwa.edu.au/">CSci UWA home</A>]['00]; uk us isbn:0864221959.

%A D. L. Dowe
%A J. Oliver
%A T. I. Dix
%A L. Allison
%A C. S. Wallace
%T A decision graph explanation of protein secondary structure prediction
%J 26th Hawaii Int. Conf. Sys. Sci.
%V 1
%P 669-678
%M JAN
%D 1993
%K LAllison, CSW, Monash, conf, MolBio, protein secondary structure prediction,
   conformation, alpha helix, ss, AA, beta sheet extended strand, turn, coil,
   II, inductive inference, decision graph tree, DTree, CSWallace, CSW,
   MML, Minimum message length encoding, description, MDL, Bayesian,
   TR163 163, c1993, c199x, c19xx, bioinformatics, HICSS, HICSS26, HICSS93
%X Oliver and Wallace (IJCAI '91) introduced `decision graphs' -
   a generalisation of decision trees - here applied to protein secondary
   structure prediction.
   [<A HREF="http://www.allisons.org/ll/Publications/1993.HICSS_26/">more</A>],
   [<A HREF="/~lloyd/tildeStrings/Protein/93.HICSS_26/93.HICSS_26.html">paper (HTML)</A>].
   Also see TR 92/163.

%A L. Allison
%A C. S. Wallace
%A C. N. Yee
%T Minimum message length encoding, evolutionary trees and multiple alignment
%J 25th Hawaii Int. Conf. on Sys. Sci.
%K LAllison, CSW, Monash, conf, MolBio, minimum message length encoding, MML,
   ML, evolutionary, family, phylogenetic, tree, trees, CSWallace, CSW, human,
   Bayesian, finite state, model, machine, FSM, hidden Markov model, primate,
   HMM, DNA, multiple alignment, inductive inference, II, bioinformatics, chimp,
   HICSS, HICSS25, HICCS92, TR 91 155, TR91-155, TR155, c1992, c199x, c19xx
%V 1
%P 663-674
%M JAN
%D 1992
%O TR 91/155, Dept. Computer Science, Monash University '91
%X "A method of Bayesian inference known as MML encoding is applied to inference
   of an evolutionary tree and to multiple alignment for K >= 2 strings.
   It allows the posterior odds-ratio of two competing hypotheses, for
   example two trees, to be calculated. A tree that is a good hypothesis forms
   the basis  of a short message describing the strings.  The mutation
   process is modelled by finite-state machine.  It is seen that tree
   inference and multiple alignment are intimately connected."
   -- [<A HREF="/~lloyd/tildeStrings/Multiple/92.HICSS_25/">paper</A>],
   there is an example on the primate globin pseudo-genes.
   (Also see [<A HREF="http://www.allisons.org/ll/Bioinformatics/Trees/">Bioinformatics</A>].)

%A L. Allison
%A C. S. Wallace
%A C. N. Yee
%T Finite-state models in the alignment of macro-molecules
%J J. Mol. Evol.
%V 35
%N 1
%P 77-89
%M JUL
%D 1992
%K LAllison, jrnl, MolBio, c1992, c199x, c19xx, TR 90/148, macromolecules,
   TR90/148, TR148, 148 inductive inference, II, DNA, bioinformatics, DPA,
   dynamic programming algorithm, mutual information, ML, string, sequence,
   comparison, alignment, minimum message length encoding, MML, FSM, FSA,
   finite state model, analysis, minimum description length, MDL, methods,
   Hidden Markov model, HMM, homology, similarity, LCS, LCSS, significance,
   evolutionary, edit distance, sequence, r-theory, linear, gap, indel, insert,
   delete, Algorithm, Time, Speed, JME, AAAI, Bayes, Bayesian, CSWallace, CSW
%O An extended abstract titled:
      Inductive inference over macro-molecules
      in joint sessions at AAAI Symposium, Stanford MAR 1990
      on  (i) Artificial Intelligence and Molecular Biology, p5-9,
      &  (ii) Theory and Application of Minimal-Length Encoding, p50-54,
   also an early version in Technical Report 90/148,
      Dept. Comp. Sci., Monash U., Australia 3168.
%X <!-- ISI j's IF=1.7 at '06 --> MML encoding is a technique of inductive inference with theoretical and
   practical advantages.  It allows the posterior odds-ratio of two theories
   or hypotheses to be calculated.  Here it is applied to the problem of
   aligning or relating two strings, in particular biological macro-molecules.
   We compare the r-theory, that the strings are related, with the null-theory,
   that they are not related. If they are related the probabilities of the
   various alignments can be calculated.  This is done for the one-, three-
   and five-state models of relation or mutation. These correspond to linear
   and piecewise linear cost functions on runs of indels.  We describe how
   to estimate the parameters of a model.  The validity of a model is itself
   a hypothesis and can be tested objectively.  This is done on real DNA
   and on artificial data. The tests on artificial data indicate limits on
   what can be inferred in various situations.  The tests on real DNA support
   either the three- or the five-state models over the one-state model.
   Finally, a fast, approximate minimum message length string comparison
   algorithm is described.
   -- [<A HREF="https://doi.org/10.1007/BF00160262">doi:10.1007/BF00160262</A>]['07].
      [<A HREF="http://www.allisons.org/ll/Publications/1992.JME/">reprint</A>] and software,
   See  C. S. Wallace  &  D.M Boulton
        An information measure for classification.
        CompJ 11(2) 185-194 Aug '68   (appendix)
        for the derivation of the coding scheme for multi-state data.
   See also (i) Bishop  &  Thompson
           (ii) Thorne,  Kishino  &  Felsenstein, and
   [<A HREF="/~lloyd/tildeStrings/Alignment/90.AIMB.ps">AIMB</A>](.ps),
   [<A HREF="http://www.allisons.org/ll/Bioinformatics/Alignment/">Alignment</A>].

%A L. Allison
%A C. S. Wallace
%A C. N. Yee
%T When is a string like a string?
%J Int. Symposium on Artificial Intelligence and Mathematics
%W Ft. Lauderdale, Florida, USA
%M JAN
%D 1990
%K LAllison, CSW, CSWallace, Monash, conf, inductive inference, II, homology,
   alignment, LCS, edit distance, string, sequence, comparison, similarity,
   r-theory, macro-molecule, MolBio, DNA, uncertainty, pattern matching, MML,
   minimum message length encoding, AIM AIM90, Hidden Markov model, HMM,
   c1990, c199x, c19xx, bioinformatics
%X [<A HREF="http://www.allisons.org/ll/Publications/1990AIM/">more</A>],
   [<A HREF="/~lloyd/tildeStrings/Alignment/90.AIM.html">html</A>],
   [<A HREF="/~lloyd/tildeStrings/Alignment/90.AIM.ps">.ps</A>](.ps)
   also see [<A HREF="/~lloyd/tildeStrings/Alignment/90.148.TR.html">TR90/148</A>](html)
        and [<A HREF="/~lloyd/tildeStrings/Alignment/90.148.TR.ps">TR90/148</A>](.ps).

%A C. S. Wallace
%T Physically random generator
%J Computer Systems Science and Engineering
%V 5
%N 2
%P 82-88
%M APR
%D 1990
%I Butterworth and Co.
%K Monash, jrnl, c1990, c199x, c19xx, random number generation, RNG,
   CSW, CSWallace
%X (Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].)

%A J. R. Neil
%A C. S. Wallace
%A K. B. Korb
%T Learning Bayesian networks with restricted causal interactions
%J UAI99
%P 486-493
%D 1999
%K conf, UAI, UAI15, UAI99, network, local models, model, c1999, c199x, c19xx,
   AI, II, stats, MML, MDL, log linear, logLinear, LLM, analysis, LLA, logit,
   cause, interaction, CSW, CSWallace
%X "A major problem for the learning of Bayesian networks (BNs) is the
   exponential number of parameters needed for conditional probability tables.
   Recent research reduces this complexity by modeling local structure in the
   probability tables. We examine the use of log-linear local models. While
   log-linear models in this context are not new (Wh'ttaker, '90;
   Bunt'ne, '91; Ne@l, '92; Heck'rman and Me'k, '97), for structure learning
   they are generally subsumed under a naive Bayes model. We describe an
   alternative interpretation, and use a Minimum Message Length (MML)
   (W@llace, '87) metric for structure learning of networks exhibiting causal
   independence, which we term first-order networks (FONs).
   We also investigate local model selection on a node-by-node basis."
   isbn:1-55860-614-9.      Auth's at Mon@sh U., .au
   [Also search for: Bayesian network].
   (Also see [<A HREF="/~lloyd/tildeImages/People/Wallace/">CSW</A>.])

%A C. S. Wallace
%A K. B. Korb
%T A Bayesian learning agent
%J Research conference: Faculty of Computing and Information Technology
%I Monash University, Melbourne
%E C. S. Wallace
%P 19
%D 1994
%K CSW, CSWallace, KBK, AI, Bayes, FCIT, inductive inference, II, machine,
   learn, Monash, c1994, c199x, c19xx
%X Also see [<A HREF="/~lloyd/tildeImages/People/Wallace/">CSW</A>].
   <A HREF="/~korb/">kbk home page</A> ('94)

%A C. S. Wallace
%A D. M. Boulton
%T An information measure for classification
%J COMPJ
%V 11
%N 2
%M AUG
%D 1968
%P 185-194
%K WB68, zzMML, CSW, CSWallace, jrnl, COMPJ, MML, MDL, II, c1968, c196x, c19xx,
   Monash, Computer Journal, SNOB, classification, class, taxonomy, cluster,
   clustering, EM algorithm, information theory, stats, IT, coding, encoding,
   minimum message length, AIC, BIC, inductive inference, minimum description
   length, probability distribution, origin, prior, multistate, multi-state,
   categorical, multinomial, complexity, binomial, normal, Gaussian, discrete,
   stochastic, algorithmic complexity, Bayes, Bayesian, Occam, Ockham, Occam's,
   razor, Rissanen, statistics, search, mixture modelling, split, merge, kill,
   EM algorithm, learning, SMEM, expectation, maximization, maximisation, model,
   citerion, seal skull, seals, skulls, Chaitin, Solomonoff, Kolmogorov,
   Figueiredo, Jain
%X "This paper derives a measure of the goodness of a classification based on
   information theory.  A classification is regarded as a method of economical
   statistical encoding of the available attribute information.
      The measure may be used to compare the relative goodness of
   classifications produced by different methods or as the basis of a
   classification procedure.
      A classification program, 'SNOB', has been written for the University
   of Sydney KDF-9 computer, and first tests show good agreement with
   conventional taxonomy.
   (An example classifies species of seal based on measurements of skulls.)"
   -- [<A HREF="http://dx.doi.org/10.1093/comjnl/11.2.185">doi:10.1093/comjnl/11.2.185</A>]['14].
   Appendix has derivation of optimal encoding for multistate data.
   [Also search for: MML].  See [<A HREF="http://www.allisons.org/ll/MML/Structured/">MML</A>],
   and [<A HREF="/~lloyd/tildeMML/Structured/1968-WB-CJ/">paper</A>].

%A D. M. Boulton
%A C. S. Wallace
%T The information content of a multistate distribution
%J J. Theor. Biol.
%V 23
%P 269-278
%D 1969
%K BW69, zzMML, CSW, CSWallace, c1969, c196x, c19xx, Minimum Message Length,
   JTB, jrnl, Encoding, MML, information, theory, IT, SC, NML, content, entropy,
   prior, adaptive code, MDL, compress, compression, BIC, string, sequence,
   multistate, multi state, multi-state, multinomial, binomial, coding,
   description, inductive inference, II, Monash, jrnl, Bayes, Bayesian, Occam,
   Occam's, Ockham razor, categorical
%X Also see Wallace & Boulton COMPJ (Computer J.) 1968.
   [<A HREF="http://dx.doi.org/10.1016/0022-5193(69)90041-1">doi:10.1016/0022-5193(69)90041-1</A>]['06],
   [<A HREF="http://www.allisons.org/ll/MML/">MML</A>].  [Also search for: MML].

%A D. M. Boulton
%A C. S. Wallace
%T An information measure for hierarchic classification
%J COMPJ
%P 254-261
%V 16
%N 3
%M AUG
%D 1973
%K BW73, CSW, Monash, jrnl, hierarchic, tree, family, class, classification,
   Minimum Message Length Encoding, MML, class, description, code, coding, MDL,
   encoding, mixture model, inductive inference, II, Computer Journal, SNOB,
   information, theory, IT, Bayes, Bayesian, Occam Occam's Ockham razor,
   c1973, c197x, c19xx, CSWallace, algorithm, extimation
%X "The information measure has been developed as a criterion of merit for
   intrinsic classifications. The information measure for non-hierarchic
   classifications has been described previously and a program developed which
   searched for that classification optimising the information measure.
   However, hierarchic classifications are often of practical importance and
   this paper develops the information measure for hierarchic classifications.
   Two algorithms are outlined for generating hierarchic classifications which
   minimise the information measure. One of these has been programmed and first
   tests show a good agreement with conventional taxonomy."
   -- [<A HREF="http://dx.doi.org/10.1093/comjnl/16.3.254">doi:10.1093/comjnl/16.3.254</A>]['05].
   Also Wallace and Boulton COMPJ (Computer J.) 1968,
   and [<A HREF="http://www.allisons.org/ll/MML/Structured/">MML</A>],
   [<A HREF="http://www.oup.co.uk/computer_journal/hdb/Volume_16/Issue_03/160254.sgm.abs.html">paper</A>][12/'03].

%A J. Patrick
%A C. S. Wallace
%T Stone circle geometries: An information theory approach
%P 231-264
%B Archaeoastronomy in the Old World
%E D. Heggie
%I CUP
%D 1982
%O Int. Symp. on Archaeoastronomy, Queen's College, Oxford, Sept. 4-9, 1981.
%K PW82, CSW, CSWallace, c1982, c198x, c19xx, Monash, inductive inference,
   II, IT, standing, megalithic, yard, stonehenge, stone circle, henge, stones,
   circles, minimum message length encoding, MML, minimum description length,
   standing, Thom, conf, chapter, MDL, Occam, Occam's, Ockham, razor, occams
%X "This article discusses the techniques of A.Thom in deriving geometric
   designs to fit stone circles & from this background argues for an alternative
   defn of an hypothesis in scientific research. The defn that is advocated
   herein is a union of Solomonoff's application of Information Theory to
   inductive inference, Wallace's Information measures & Halstead's software
   science measures. This approach is applied to the comparison of Thom's
   hypothesis against the authors' hypothesis that stone circles are meant to be
   roughly circular & locally smooth to the eye. The authors' hypothesis is
   modelled by a fourier series wrapped around a circle. The results from 65
   Irish sites show that the authors' hypothesis is favoured at odds of better
   than 780:1 compared to Thom's hypothesis."
   -- [<A HREF="http://dx.doi.org/10.1017/CBO9780511898310.016">doi:10.1017/CBO9780511898310.016</A>][11/'11].
   (In: hb us$52, uk us isbn:0521125308, uk us isbn13:978-0521125307.)
   (Also see [<A HREF="http://www.allisons.org/ll/MML/">MML</A>].)
   [Also search for: megalithic stone circle].

%A C. S. Wallace
%T On the identification of outliers in a simple model
%I Dept. Computer Science, Monash University
%D 1984 ?
%K CSWallace, CSW, c1984, c198x, c19xx, description of outliers, outlier, stats,
   statistics, minimum message length, MML, MDL, noise, noisy, data, discordancy
%X "We suppose that we are given a set of 'N' observations {yi, i=1,...,N} which
   are thought to arise independently from some process of known form & unknown
   (vector) parameter theta. However, we may have reason to suspect that some
   small fraction of the N obs. are in some sense contaminated or erroneous,
   i.e., that they arise from a process different from the main process.
   Any such obs. is called an "outlier". We will then be interested in methods
   for identifying or at least estimating the # of the outliers, & for
   estimating theta in a way which is minimally upset by the outliers. . . ."
   -- [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/Outlier/">More (click)</A>].
   (Written in the 1980s.)

%A M. P. Georgeff
%A C. S. Wallace
%T A general selection criterion for inductive inference
%J European Conf. on Artificial Intelligence
%P 473-482
%W Pisa
%M SEP
%D 1984
%I Elsevier Sci. Publ.
%C New York
%O TR 32, CS Monash March 1983
%K GW84, zzMML, CSW, conf, ECCAI, ECCAI84, information theory, IT, II, Bayesian,
   language, minimum message length, MML, encoding, description, Occam, Ockham,
   curve, Occam's razor, MDL, simplicity, hypothesis, line, lines, fit, fitting,
   infer, inferring, string, strings, probabilistic finite state automaton,
   automata, PFSA, FSA, Hidden Markov Models, HMM, TR32, TR 32,
   c1984, c198x, c19xx, complexity, Kolmogorov, Solomonoff, AIC, CSWallace
%X The nature of theories, structural descriptions,
   inference of line segments,
   There are more examples, including inference of probabilistic
   finite state automata (PFSA) for languages in the technical report.
   -- [<A HREF="http://www.allisons.org/ll/MML/Structured/TR32/">TR 32</A>],
   or [<A HREF="/~lloyd/tildeMML/Structured/TR32/">TR 32</A>].

%A C. S. Wallace
%A P. R. Freeman
%T Estimation and inference by compact coding
%J J. of the Royal Statistical Soc. series B
%V 49
%N 3
%D 1987
%P 240-265
%K WF87, zzMML, CSW, CSWallace, Monash, jrnl, JRSS, RSS, JRSSB, MML87, stats,
   strict minimum message length encoding, MML, SMML, inductive inference, II,
   information theory, IT, minimum description length, MDL, Jorma Rissanen,
   SNOB, mixture model, modelling, classification, class, classification,
   cluster, clustering, probability distribution, normal, stochastic, prior,
   Bayes, Bayesian, Occam, Occam's, Occams, Ockham, razor, algorithm,
   estimation, algorithmic complexity, AC, AIC, BIC, Fisher information,
   Kolmogorov, c1987, c198x, c19xx
%X "The systematic variation within a set of data, as represented by a usual
   stat. model, may be used to encode the data in a more compact form than would
   be possible if they were considered to be purely random. The encoded form has
   two parts. The 1st states the inferred estimates of the unknown parameters
   in the model, the 2nd states the data using an optimal code based on the data
   prob. distn implied by those param. estimates. Choosing the model & the
   estimates that give the most compact coding leads to an interesting general
   inference procedure. In its strict form it has great generality & several
   nice properties but is computationally infeasible. An approx. form is
   developed and its relation to other methods is explored."
   -- There was a special session of the Royal Stat. Soc. and
   Vol.49 No.3 pp.223-265 includes  Rissanen 's paper, W + F, and discussion.
   -- [<A HREF="http://www.allisons.org/ll/MML/Structured/1987-WF-JRSS.html">paper</A>],
      [<A HREF="http://www.allisons.org/ll/MML/">MML</A>],
   or [<A HREF="/~lloyd/tildeMML/Structured/1987-WF-JRSS.html">paper</A>],
   2985992@[<A HREF="http://www.jstor.org/stable/2985992">Jstor</A>]['14].

%A C. S. Wallace
%T Classification by minimum-message-length encoding
%B Advances in Computing and Information - ICCI '90
%I SpringerVerlag
%S LNCS
%V 468
%P 72-81
%M MAY
%D 1990
%K Wal90, zzMML, CSW, Monash, conf, ICCI 90, ICCI90, inductive inference, II,
   SNOB, class, classify, classes, classification, cluster, clustering, mixture,
   model, modelling, c1990, c199x, c19xx, minimum message length encoding, MML,
   description, MDL, dld, bits back, mmld, Bayes, Bayesian, Occam, Ockham razor,
   Snob2, fractional, soft assignment, bit borrowing, Hinton, Camp, CSWallace
%X 'Although classification is perhaps the oldest practical application of MML
   inference, the early algorithm was subject to weakly inconsistent estimation.
   The same problem is inherent in any MML inference which infers many discrete
   "nuisance" parameters. A solution has been found using a novel coding trick,
   which could be useful in many inductive inferences.'
   -- [<A HREF="http://dx.doi.org/10.1007/3-540-53504-7_63">doi:10.1007/3-540-53504-7_63</A>]['11]; uk us isbn:3540535047.
   Note, sec.3 "partial assignment," p.75, includes a discussion of
      partial- / fractional- / soft-assignment of a datum to a cluster (class).
   And on p.81,
      "... a more detailed descn of the program is given by ...
      Wallace, 'An improved program for classification',
      Monash U., Comp. Sci. TR 47 (1984)."
         {It actually says by "Wallace & Patrick" but TR47 (1984) is
         actually by Wallace alone; I have a copy -- L.A. 16/7/'14.}
   [Also search for: Wallace 1986 ACSC9], and
   see [<A HREF="http://www.allisons.org/ll/MML/Structured/">MML</A>].

%A J. J. Oliver
%A C. S. Wallace
%T Inferring decision graphs
%J IJCAI-91, workshop 8
%P ???-???
%M JAN
%D 1991
%O TR 92/170, Dept Computer Science, Monash U.
%K OW91, CSW, IJCAI, IJCAI91, MML, minimum message length encoding,
   classification, inductive inference, II, machine learning,
   disjunctive decision tree DTree, DGraph, graph graphs, description, DAG,
   Monash, conf, MDL, c1991, c199x, c19xx, Bayes, Bayesian,
   Occam Occam's Ockham razor, CSWallace, TR 170, TR170
%X (Also see [<A HREF="http://www.allisons.org/ll/MML/Structured/DGraph/">MML</A>].)
   [Also search for: Dgraph]  and  [also search for: Dtree].

%A C. S. Wallace
%A J. D. Patrick
%T Coding decision trees
%J Machine Learning
%V 11
%P 7-22
%D 1993
%K WP93, CSW, CSWallace, c1993, c199x, c19xx, Monash, DTree, decision tree,
   trees, Quinlan, Rivest, classification, minimum message length encoding,
   MML, description, classification tree, CTree, MDL, inductive inference, II,
   jrnl, TR 91 153 TR91/153 TR153, information theory, Bayes, Bayesian, Occam,
   Occam's, Ockham, razor, code
%O TR 91/153, Comp. Sci., Monash U.
%X "Quinlan & Rivest have suggested a [D]-tree inference method using the
   [MDL] idea. We show that there is an error in their derivation of message
   lengths, which fortunately has no effect on the final inference. We further
   suggest 2 improvements to their coding techniques, one removing an
   inefficiency in the description of non-binary trees, & one improving the
   coding of leaves. We argue that these improvements are superior to similarly
   motivated proposals in the original paper.
   Empirical tests confirm the good results reported by Quinlan & Rivest, &
   show our coding proposals to lead to useful improvements in the performance
   of the method."
   -- [<A HREF="http://dx.doi.org/10.1023/A:1022646101185">doi:10.1023/A:1022646101185</A>][8/'04].
   [Also search for: decision graph]  with J. O l i v e r,  and
   see [<A HREF="http://www.allisons.org/ll/MML/Structured/DTree/">MML</A>].
   (Also see, sec.2.1.14, sec.2.1.15 & sec.2.1.16 of [CSWallace book], 2005,
   on codes for integers.)

%A C. S. Wallace
%A P. R. Freeman
%T Single-factor analysis by minimum message length estimation
%J J. of the Royal Statistical Soc. series B
%V 54
%N 1
%P 195-209
%D 1992
%K WF92, zzMML, CSW, CSWallace, jrnl, JRSS, RSS, JRSSB, single factor analysis,
   FactorAnalaysis, mixture model, classification, FA, MML,
   minimum description length, encoding, inductive inference, II, multivariate,
   estimate, stats, statistics, MDL, nuisance, hidden, factors, load, loadings,
   Bayes, Bayesian, Occam, Occam's, Monash, SNOB, Ockham razor,
   dimension reduction, c1992, c199x, c19xx
%X "The minimum message length (MML) technique is applied to the problem of
   estimating the parameters of a multivariate Gaussian model in which the
   correlation structure is modelled by a single common factor. Implicit
   estimator equations are derived and compared with those obtained from a
   maximum likelihood (ML) analysis. Unlike ML, the MML estimators remain
   consistent when used to estimate both the factor loadings and the factor
   scores. Tests on simulated data show the MML estimates to be on average more
   accurate than the ML estimates when the former exist. If the data show
   little evidence for a factor, the MML estimate collapses. It is shown that
   the condition for the existence of an MML estimate is essentially that the
   log-likelihood ratio in favour of the factor model exceeds the value
   expected under the null (no-factor) hypotheses."
   Also see [<A HREF="http://www.allisons.org/ll/MML/Notes/SNOB-factor/">factor-snob</A>],
      [<A HREF="/~lloyd/tildeMML/Structured/1992-WF-JRSS.html">paper</A>],
      2345956@[<A HREF="http://www.jstor.org/stable/2345956">Jstor</A>]['14].
   [Also search for: MML multiple factor analysis].

%A C. S. Wallace
%T Multiple factor analysis by MML estimation
%R 95/218
%I Dept. Comp. Sci., Monash U.
%M MAR
%D 1995
%K Wal95, zzMML, CSW, CSWallace, TR 218, TR218, minimum message length, MML,
   FA, FactorAnalysis, description, multiple factor analysis, factors, MDL,
   inductive inference, II, statistics, stats, factors, dimension,
   dimensionality reduction, Monash, c1995, c199x, c19xx
%X Also see (i) the single factor paper:  JRSS(B) 54(1), pp.195-209, '92.
   and (ii) Proc. Bien' Stat' Conf',  Brisbane, '98.
   Ordering TR 95/218: Faculty of Info. Tech. (Clayton), Monash Uni., .au 3800.
   Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/Multi-Factor/">multi-F</A>],
   and [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/FactorSnob/">factor Snob</A>].

%A C. S. Wallace
%T False oracles and SMML estimators
%R TR 128
%I Dept. Computer Science, Monash University
%D 1989
%K Wal89, zzMML, CSW, CSWallace, TR 128, TR128, oracle, c1989, c198x, c19xx,
   strict minimum message length, encoding, MML, inductive inference, II, BIC,
   description, information theory, IT, foundations, AIC, AICc, MDL, oracle,
   Monash, csse, point, estimator, estimation, estimate, Bayesian, theoretical,
   Kolmogorov complexity, SMML, foundations, statistics, stats, methodology
%X "An attempt is made to formalize the notion of the empirical comparison
   of two estimators for the same problem. By "empirical", we mean that the
   comparison be based on the estimates produced by the estimator for some set
   of data, not on how they produce them. Defining the `oracle' as an estimator
   which gives the true values of the unknown parameters, we define `false
   oracle' as an estimator which no fair empirical criterion can be expected to
   reject in favour of the oracle, and argue that false oracles exist, but are
   stochastic functions of the data. Further, we show that Strict Minimum
   Message Length estimators closely imitate false oracles despite being
   deterministic functions. Some practical consequences are exhibited."
   (Ordering TR89/128: Faculty of Info. Tech. (Clayton), Monash Uni., .au 3800.)
   [Also search for: ISIS96] and see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/False-Oracles/">False-Oracles</A>].

%A D. P. McKenzie
%A P. D. McGorry
%A C. S. Wallace
%T Constructing minimal diagnostic decision trees
%J Australian Soc. for Psychiatric Res. Conf., Adelaide.
%D 1992
%K conf, CSWallace, Monash, MML, MDL, DTree, tree, c1992, c199x, c19xx
%X Poster. [Also search for: McKenzie c1993]
   See [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A D. P. McKenzie
%A P. D. McGorry
%A C. S. Wallace
%A L. H. Low
%A D. L. Copolov
%A B. S. Singh
%T Constructing a minimal length diagnostic decision tree
%J Methods in Information in Medicine.
%V 32
%N 2
%P 161-166
%M APR
%D 1993
%K CSW, CSWallace, Monash, supervised classification jrnl, c1993, c199x, c19xx,
   DTree, trees, minimum message length encoding, MML, inductive inference, II,
   Bayes, Bayesian, calssification, medical, diagnose, description, MDL
%X "CTs & discriminant function analysis ... to ascertain whether a small # of
   diagnostic decision rules could be extracted from a large inventory of items.
   Several models, involving up to 17 symptoms, that led to a broad psychiatric
   diagnosis were then tested on a small validation sample of 53 patients. All
   methods, with the exception of CART used without any pruning, generated
   identical trees involving four items. Almost 90% of the validation sample
   was able to be correctly classified by all methods although poor
   classification performance was noted in the case of 1 particular diagnosis,
   Schizoaffective Psychosis. In contrast, stepwise LDA originally selected 17
   items, although 3 out of the first 4 items selected were identical to those
   chosen by the tree-building methods. Although more research is required,
   there are indications that the latter methods may be usefully employed in
   constructing parsimonious decision trees."
   Also see [<A HREF="http://www.allisons.org/ll/MML/">MML</A>].

%A C. S. Wallace
%A D. L. Dowe
%T Estimation of the von Mises concentration parameter using minimum message
   length
%J Proc. 12th. Aust. Stat. Soc. Conf.
%P ??-??
%D 1994
%K CSW, DLD, Monash, conf, vonMises, minimum message length encoding, MML,
   Bayes, Bayesian, SNOB, class, classify, classification, cluster, circle,
   circular, probability distribution, angle, direction, description,
   CSWallace, TR193, TR 193, zz0795, c1994, c199x, c19xx
%O W & D,
     'MML estimation of the von Mises concentration parameter,' TR 93/193,
     Dept. Comp. Sci., Monash Uni., Clayton 3168, Australia, pp.20,  1993;
   also
   D, O, B & W,
     'Bayesian estimation of the Von Mises concentration parameter',
     Proc. 15th Int. Wkshop on Max Ent. and B. Methods, Santa Fe, '95
%X (Was [<A HREF="ftp://ftp.csse.monash.edu.au/publications/1993/TR193.ps.gz">TR193</A>]['94].)
   [Also search for: vonMises direction], and
   [also search for: probability distribution circle], and
   also see [<A HREF="http://www.allisons.org/ll/MML/">MML</A>].

%A C. S. Wallace
%A D. L. Dowe
%T Intrinsic classification by MML - the Snob program
%J Proc. 7th Australian Joint Conf. on Artificial Intelligence
%W Armidale, NSW, Australia
%I World Scientific
%P 37-44
%M NOV
%D 1994
%K AJCAI, AJCAI7, ACAI7, AI, CSW, CSWallace, DLD, monash, conf, class,
   clustering, cluster, inductive inference, SNOB, c1994, c199x, c19xx,
   minimum message length MML, description, encoding, MDL, von Mises, vonMises
%X (Also see [<A HREF="http://www.allisons.org/ll/MML/">MML</A>].)

%A C. S. Wallace
%T Multiple factor analysis by MML estimation
%J Proc. of the 14th Biennial Statistical Conf.
%W Brisbane
%P 144
%M JUL
%D 1998
%K Wal98, zzMML, CSW, CSWallace, minimum message length, MML, description,
   c1998, c199x, c19xx, factors, MDL, MFA, multiple factor, multivariate,
   Gaussian, normal, AIC, maximum likelihood, conf, common, score, scores,
   FactorAnalysis, dimension, dimensionality reduction,
   ratio, test, tests, stats, statistics, FA
%X "... The [MML] estimator has been extensively tested and compared to the
   maximum likelihood and AIC estimators.  The MML estimator is found to be
   substantially more accurate, to provide consistent estimates of the factor
   scores and to recover the number of common factors more reliably than a
   likelihood ratio test among maximum likelihood models."
   Also see (i) the single-factor paper JRSS 1992, and (ii) TR 95/218,
   and [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/Multi-Factor/">MML-factors</A>].

%A D. W. Kissane
%A S. Bloch
%A W. I. Burns
%A J. D. Patrick
%A C. S. Wallace
%A D. P. McKenzie
%T Perceptions of family functioning and cancer
%J Psycho-oncology
%V 3
%P 259-269
%D 1994
%K jrnl, psychooncology, medical, minimum message length, MML, description, MDL,
   application, grief, CSW, CSWallace, c1994, c199x, c19xx, psychology,
   families
%X "Family function was assessed in 102 families (342 members) of palliative
   care patients & grouped into classes by a computer-based taxonomic program.
   Five classes were defined through the dimensions of cohesiveness, conflict &
   expressiveness of the Family Environment Scale (FES). One third of families
   we named supportive for their high cohesiveness; a further 21% resolved
   conflict effectively; both of these classes contained low psychological
   morbidity. Two classes (15%) were clearly dysfunctional: hostile families
   (6%) were distinguished by high conflict while sullen families (9%) displayed
   moderate conflict, poor cohesion & limited expressiveness. These two classes
   had significantly higher levels of psychological morbidity & poorer social
   functioning. The remaining class (31%) had intermediate levels of cohesion,
   expressiveness & conflict (termed ordinary) yet more moderate psychosocial
   morbidity. Screening of families with the FES would facilitate a more
   family-centred approach to treatment, with relatively early identification
   of families at-risk; preventive interventions would also then be feasible."
   -- [<A HREF="http://dx.doi.org/10.1002/pon.2960030403">doi:10.1002/pon.2960030403</A>]['14].
   Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A D. W. Kissane
%A S. Bloch
%A D. L. Dowe
%A R. D. Snyder
%A P. Onghena
%A D. P. McKenzie
%A C. S. Wallace
%T The Melbourne family grief study, I: Perceptions of family functioning
   in bereavement
%J American Journal of Psychiatry
%V 153
%P 650-658
%D 1996
%K DLD, CSW, CSWallace, Monash, jrnl, AJP, psychiatry, psychology, grief,
   application, SNOB, application, MML, II, c1996, c199x, c19xx
%X Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A J. J. Oliver
%A R. A. Baxter
%A C. S. Wallace
%T Unsupervised learning using MML
%J Machine Learning: Proc. 13th Int. Conf. (ICML 96)
%E L. Saitta
%I MorganKaufmann
%P 364-372
%D 1996
%K conf, ICML96, ICML13, minimum message length, MML, description, MDL, Snob,
   classification, model, multivariate, stats, statistics, inductive inference,
   CSW, CSWallace, clustering, II, AI, zz1296, c1996, c199x, c19xx
%X Snob has been widely used for mixture modelling and uses the minimum
   message length criterion. The message length formulas in Snob are
   based on early work. This paper updates the message length formula
   based on general recent work.  The MML criterion is compared to other
   prominent criterion in a series of Monte Carlo experiments.  (via rohan)
   [Also search for: MML],  and
   see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A J. J. Oliver
%A R. A. Baxter
%A C. S. Wallace
%T Minimum message length segmentation
%J Res. and Dev. in Knowledge Discovery and Data Mining (PAKDD-98)
%P 222-233
%D 1998
%K conf, segment, cut point, points, cutpoint, MML, MDL, series approximation,
   jono, CSW, CSWallace, Monash, c1998, c199x, c19xx
%X Also see:  Baxter Oliver c1996.
   And [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A D. L. Dowe
%A R. A. Baxter
%A J. J. Oliver
%A C. S. Wallace
%T Point estimation using the Kullback-Leibler loss function and MML
%J 2nd Pacific-Asia Conf. on Knowledge Discovery and Data Mining (PAKDD98)
%S LNAI
%V 1394
%P 87-95
%M APR
%D 1998
%K conf, PAKDD, PAKDD2, CSW, CSWallace, MML, MDL, KL, c1998, c199x, c19xx
%X Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T False oracles and SMML estimators
%J Proc. Int. Conf. on Information, Statistics and Induction in Science
%I World Scientific
%P 304-316
%D 1996
%K Wal96, zzMML, conf, ISIS, ISI96, strict, minimum message length encoding,
   SMML, MML, inductive inference, II, description, information theory, IT,
   foundations, MDL, statistics, Bayes, Bayesian, estimate, AI,
   CSW, CSWallace, c1996, c199x, c19xx, zz1296, TR 128, TR128
%O TR 89/128, Dept. Comp. Sci., Monash Univ., '89
%X "MML and MDL inductive inference techniques are based on the information-
   -theoretic notion of transmitting data concisely. Both methods are
   universally applicable, consistent and efficient. MML [LA: ie MML87] is a
   Bayesian technique and is invariant under parameter transformations; however
   it is a quadratic approximation to a slightly more efficient two-part coding
   technique, Strict MML (SMML), which maps from the data space onto a
   countable, discretized subset of the parameter space. It is shown that the
   posterior distribution of the parameter vector is a "false oracle" in that
   no fair comparison between the true ("oracular") parameter vector and a
   sampling from the posterior will enable us to distinguish one from the other.
   It is further shown that the invariant, consistent and efficient Bayesian
   SMML point estimation technique closely approximates (and converges to) a
   false oracle. Hence, SMML inductions are practically indistiguishable from
   the truth in the absence of data other than that used in the induction."
   -- [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/False-Oracles/">paper@www</A>].
   [Also search for: TR128 CSW].

%A C. S. Wallace
%A D. L. Dowe
%T MML mixture modelling of multi-state, Poisson, von Mises circular
   and Gaussian distributions
%J Proc 28th Symp. on the Interface
%P 608-613
%D 1997
%O Preliminary papers of the 6th Int Workshop on AI and Stats,
      Ft. Lauderdale, Florida, 529-536, Jan 1997
%K MML, MDL, inductive inference, II, clustering, classification, class,
   cluster, normal, periodic, Snob, conf, AIS6, AIS97, CSW, DLD,
   vonMises, zz0197, CSWallace, c1997, c199x, c19xx
%X [Also search for: MML SNOB]  and  [also search for: vonMises direction]
   and see [<A HREF="http://www.allisons.org/ll/MML/">MML</A>].

%A D. L. Dowe
%A J. J. Oliver
%A C. S. Wallace
%T MML estimation of the parameters of the spherical Fisher distribution
%J Proc. 7th Int w'shop on Alg. Learning Complexity
%D 1996
%W Sydney
%S LNAI
%N 1160
%P 213-227
%O TR 97/272, Dept. Comp Sci. Monash University
%K ALT, ALT96, ALT7, conf, MML, MDL, probability, DLD, CSW, CSWallace,
   c1996, c199x, c19xx
%X [Also search for: MML vonMises].
   (Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].)

%A D. L. Dowe
%A C. S. Wallace
%T Resolving the Neyman-Scott problem by Minimum Message Length (abstract)
%J Proc. Sydney Int. Stat. Congress
%P 197-198
%D 1996
%O TR 97/307, Department of Computer Science, Monash University, Australia,
   Feb 1997
%K abstract, conf, SISC, SISC96, TR 307, TR307, MML, description, MDL, DLD, CSW,
   CSWallace, c1996, c199x, c19xx
%X authors at Comp. Sci., Monash University, Vic, oz
   also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T MML inference of predictive trees, graphs and nets
%B Computational Learning and Probabilistic Reasoning
%E A. Gammerman
%I Wiley
%P 43-66
%D 1996
%K minimum message length, MML, MDL, description, DTree, c1996, c199x, c19xx,
   inductive inference, II, CSW, CSWallace, chapter
%X isbn:0-471-96279-1.
   Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T Intrinsic classification of spatially correlated data
%J COMPJ
%V 41
%N 8
%P 602-611
%D 1998
%K jrnl, COMPJ, cluster, class, clustering, classification, region, geographic,
   classify, data mining, mine, knowledge discovery, KD, DM, AI, HMM, Markov,
   random field, mixture modelling, spatial, fields, model, pattern, 2D, 3D,
   Monte Carlo, MCMC, Gibbs sampling, minimum message length encoding, MML, EM,
   SNOB, correlation, Markov, inductive inference, minimum description length,
   II, zz0499, c1998, c199x, c19xx, MDL, CSWallace, CSW, Monash
%X "Intrinsic classification, or unsupervised learning of a classification,
   was the earliest application of what is now termed minimum message length
   (MML) or minimum description length (MDL) inference. The MML algorithm
   Snob[1] and its relatives have been used successfully in many domains. These
   algorithms treat the things to be classified as independent random
   selections from an unknown population whose class structure, if any, is to
   be estimated. This work extends MML classification to domains where the
   things have a known spatial arrangement and it may be expected that the
   classes of neighbouring things are correlated. Two cases are considered.
   In the first, the things are arranged in a sequence and the correlation
   between the classes of successive things modelled by a first-order Markov
   process. An algorithm for this case is constructed by combining the Snob
   algorithm with a simple dynamic programming algorithm. The method has been
   applied to the classification of protein secondary structure. In the second
   case, the things are arranged on a two-dimensional (2D) square grid, like
   the pixels of an image. Correlation is modelled by a prior over patterns
   of class assignments whose log probability depends on the number of adjacent
   mismatched pixel pairs. The algorithm uses Gibbs sampling from the pattern
   posterior & a thermodynamic relation to calculate message length." [17 refs]
   -- [<A HREF="http://dx.doi.org/10.1093/comjnl/41.8.602">doi:10.1093/comjnl/41.8.602</A>]['05].
   [Also search for: Edgoose DCC 1998], and
   see [<A HREF="http://www.allisons.org/ll/MML/">MML</A>],
   and [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A M. Viswanathan
%A C. S. Wallace
%T A note on the comparison of polynomial selection methods
%J Proc. 7th Int. Workshop on Artif. Intell. and Stats.
%W Ft. Lauderdale
%M JAN
%D 1999
%P 169-177
%I MorganKauffman
%K conf, fitting, basis functions, minimum message length, description, MDL,
   Murli, CSW, CSWallace, MML, SVM, zz1099, c1999, c199x, c19xx, curve, fit,
   regression, method, structural risk minimisation, SRM, minimization, AIS
%X "Minimum Message Length (MML) & Structural Risk Minimisation (SRM) are 2
   computational learning principles that have achieved wide acclaim in recent
   years. ...former is based on Bayesian learning & the latter on the classical
   theory of VC-dimension, they are similar in their attempt to define a
   trade-off between model complexity & goodness of fit to the data. A recent
   empirical study by Wallace compared the perf. of std model sel. methods in a
   1-D polynomial regression framework. ... strong evidence in support of the
   MML & SRM based methods over the other std approaches. [here] present a
   detailed empirical evaluation of 3 model sel. methods which include an MML
   based approach & 2 SRM based methods. ... suggest that the MML-based approach
   ...has higher predictive accuracy & also raise questions on the inductive
   capabilities of [SRM]."  [11 Refs].
   (Also see [<A HREF="http://www.allisons.org/ll/MML/">MML</A>].)

%A M. Viswanathan
%A C. S. Wallace
%A D. L. Dowe
%A K. B. Korb
%T Finding cutpoints in noisy binary sequences - a revised empirical evaluation
%J Proc 12th Australian Joint Conf. on Artif. Intell.
%W Melbourne
%P 405-416
%D 1999
%K conf, AI, cut point, points, segmentation, MML, MDL, minimum message length,
   Kearns, series, Murli, CSW, CSWallace, DLD, description, zz1099, segment,
   c1999, c199x, c19xx
%X ... Kearns et al ... risk minimization, MDL, cross validation, ...
   inefficiency in the [original] MDL approach, ... give revised MDL method &
   another based on minimum message length (MML) principle.  [8 Refs].
   -- [<A HREF="http://dx.doi.org/10.1007/3-540-46695-9_34">doi:10.1007/3-540-46695-9_34</A>]['09].
   (Also see [<A HREF="http://www.allisons.org/ll/MML/">MML</A>].)

%A G. E. Farr
%A C. S. Wallace
%T Algorithmic and combinatorial problems in strict minimum message length
   inference
%J Res. in Combinatorial Algorithms
%P 50-58
%D 1997
%K jrnl, MML, SMML, binomial, trinomial, multinomial, probability distribution,
   inductive inference, II, AI, CSW, CSWallace, c1997, c199x, c19xx, zz0801
%X Construction of the SMML estimator is NP-hard for most distributions.
   The only two dist' for which it has been constructed are
   binomial and trinomial.
   Also see: Farr and Wallace, Comp. Jrnl., Vol.45(3), '02,
   and [<A HREF="http://www.allisons.org/ll/MML/">MML</A>],
   or [also search for: MML].

%A G. W. Rumantir
%A C. S. Wallace
%T Sampling highly correlated data for polynomial regression and model discovery
%J IDA 2001
%S SpringerVerlag
%S LNCS
%V 2189
%P 370-377
%M SEP
%D 2001
%K conf, IDA01, IDA2001, polynomials, regression, inductive inference, II,
   Intelligent Data Analysis, minimum message length, MML, description, MDL,
   GWR, GWRumantir, CSW, CSWallace, c2001, c200x, c20xx, zz0202, models
%X "The usual way of conducting empirical comparisons among competing polynomial
   model selection criteria is by generating artificial data from created true
   models with specified link weights. The robustness of each model seln
   criterion is then judged by its ability to recover the true model from its
   sample data sets with varying sizes and degrees of noise.
   If we have a set of multivariate real data & have empirically found a poly.
   regression model that is so far seen as the right model represented by the
   data, we would like to be able to replicate the multivariate data
   artificially to enable us to run multiple experiments to achieve two
   objectives. First, to see if the model selection criteria can recover the
   model that is seen to be the right model. Second, to find out the min. sample
   size required to recover the right model. ...
   proposes a methodology to replicate real multivariate data using its
   covariance matrix and a polynomial regression model seen as the right model
   represented by the data. The sample data sets generated are then used for
   model discovery experiments."
   -- [<A HREF="http://dx.doi.org/10.1007/3-540-44816-0_37">doi:10.1007/3-540-44816-0_37</A>]['11].
   Author at  M o n a s h  uni  2/'02.
   (Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].)

%A G. W. Rumantir
%A C. S. Wallace
%T Minimum message length criterion for second-order polynomial model selection
   applied to tropical cyclone intensity forecasting
%J Int. Symp. Intell. Data Analysis, IDA 2003
%P 486-496
%I SpringerVerlag
%S LNCS
%V 2810
%D 2003
%K conf, IDA, IDA2003, CSW, CSWallace, MML, MDL, models, c2003, c200x, c20xx,
   weather, climate, regression, II, AI, minimum message length, description,
   AIC, SRM, SC, forecast, cyclones, hurricane, storm, strength
%X "... tries to merge polynomial model seln research & tropical cyclone
   forecasting res.. The contributions of the work are 4-fold. First, a new
   criterion based on the [MML] principle specifically formulated for the task
   of polynomial model seln up to the 2nd order is presented. Second, a
   programmed optimisation search alg. for 2nd-order polynomial models that can
   be used in conjunction with any model selection criterion is developed.
   Third, critical examinations of the differences in performance of the various
   criteria ... . Fourth, a novel strategy which uses a synergy between the new
   criterion built based on the MML principle & other model seln criteria namely
   [MDL], Corrected Akaike's Information Criterion, Structured Risk Minimization
   & Stochastic Complexity is proposed. The forecasting model developed using
   this new automated strategy has better performance than the benchmark models
   SHIFOR (Statistical HurrIcane FORcasting) & SHIFOR94 ..."
   -- [<A HREF="http://dx.doi.org/10.1007/978-3-540-45231-7_45">doi:10.1007/978-3-540-45231-7_45</A>]['11].
      [Also search for: polynomial regression] and
      [also search for: polynomial MML].
   (Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].)

%A M. Viswanathan
%A C. S. Wallace
%T An optimal approach to mining Boolean functions from noisy data
%J Intelligent Data Engineering and Automated Learning
%P 717-724
%I SpringerVerlag
%S LNCS
%V 2690
%M SEP
%D 2003
%O 4th Int. Conf. Intelligent Data Engineering and Automated Learning,
   IDEAL 2003, Hong Kong, China, March 21-23
%K minimum message length, MML, segmentation, description, CSW, CSWallace, MDL,
   Kearns, Murli, boolean, binary function, guaranteed risk minimization,
   conf, IDEAL, AI, II, cross validation, c2003, c200x, c20xx
%X "Data Mining of binary seqs. ... The inference task in this paper, a
   specialized version of the segmentation problem, is the estimation of a
   predefined Boolean function on the real interval [0,1] from a noisy random
   sample. The framework for this problem was introduced by Kearns et al (1997)
   .... This paper presents an optimal approach to mining for Boolean fns from
   noisy data samples based on the [MML] principle. [It] is shown to be optimal
   in comparison to well-known model selection methods based on Guaranteed Risk
   Minimization, Minimum Description Length (MDL) Principle & Cross Validation
   after a thorough empirical evaluation with varying levels of noisy data."
   [<A HREF="http://www.springerlink.com/openurl.asp?genre=article&issn=0302-9743&volume=2690&spage=717">openurl</A>]['04],  isbn:354040550X.
   Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A M. Anderson
%A R. D. Pose
%A C. S. Wallace
%T A password capability system
%J COMPJ
%V 29
%N 1
%M Feb
%D 1986
%P 1-8
%K jrnl, COMPJ, Monash, CSW, CSWallace, multiprocessor, multi processor, OS,
   operating system, capabilities, c1986, c198x, c19xx, concurrent, money,
   charging
%X A tightly coupled multiprocessor.
   supports a uniform virtual memory
   unusual capability mechanism and the introduction of a money mechanism.
   simple and flexible solution to the confinement problem.
   -- [<A HREF="http://dx.doi.org/10.1093/comjnl/29.1.1">doi:10.1093/comjnl/29.1.1</A>]['05],
   CSW's [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">publns</A>].

%A C. S. Wallace
%T Statistical and Inductive Inference by Minimum Message Length
%I SpringerVerlag
%S Information Science and Statistics
%P 432
%M MAY
%D 2005
%K Wal05, W05, zzMML, c2005, c200x, c20xx, CSW, CSWallace, Chris, CSSE, Monash,
   book, text, description, MML, II, AI, AIC,BIC,NML, stats, theory, MDL, Snob,
   machine learning, compression, compact coding, Kolmogorov complexity,
   algorithmic complexity, Bayesian, model, models, selection, MML87, SMML,
   MMLa, CSci, L4ref, L5ref, Occam's, Ockham's razor, information theory, IT,
   criterion, encoding, coding, uncertainty, estimate, estimation, estimator,
   theoretical, stochastic, statistics, entropy, minimum message length,
   feathers on the arrow of time
%X "Since 1965, Prof. Wallace & others have been developing an approach
   to statistical estimation, hypothesis testing, model selection & their
   applications in the [AI] field of Machine Learning. The approach is based
   on Information Theory, using concepts from classical Shannon theory & more
   recent work on Algorithmic Complexity. The new approach has come to be
   called the Minimum Message Length principle, since it is based on the idea
   of constructing a message which concisely encodes the available data.
   Although a range of jrnl & conf. papers has been published on the principle
   & its application, & several computer programs applying it have been shown
   to perform well & have been fairly widely used, there is no text providing
   a thorough treatment of the principle or giving general guidance for its
   application."  [-- until now.]
   1st ed 2005; hb us$80; uk us isbn:038723795X; uk us isbn13:9780387237954;
   [<A HREF="http://www.allisons.org/ll/MML/2005book.html">book-page</A>][10/'04],
   [<A HREF="https://doi.org/10.1007/0-387-27656-4">doi:10.1007/0-387-27656-4</A>]['17].
   (Note, Chapter_8, 'The Feathers on the Arrow of Time.')
   (Also see [<A HREF="http://www.allisons.org/ll/MML/">MML</A>], and
   the [<A HREF="http://www.allisons.org/ll/MML/Entropy/Gas/">Ideal_Gas</A>].)

%A C. S. Wallace
%A M. B. Dale
%T Hierarchical clusters of vegetation types
%J Commun. Ecol.
%I Akademiai Kiado
%V 6
%N 1
%P 57-74
%M JUN
%D 2005
%K jrnl, CSW, CSWallace, MBDale, c2005, c200x, c20xx, ecology, type, cluster,
   Minimum Message Length, MML, description, MDL, stats, hierarchic,
   clustering, mixture model, classification, Snob
%X "... a hierarchical solution was not as efficient as a nonhierarchical one.
   However, the hierarchical solution seems to provide a more comprehensible
   solution, separating first isolated types, probably caused from unusual
   contingent events, then subdividing the more diverse areas before finally
   subdividing the less diverse. By presenting this in 3 stages, the complexity
   of the non-hierarchical result is avoided. The result also suggests that a
   hierarchical analysis may be useful in determining homogeneous areas."
   issn:1585-8553.
   Also see [<A HREF="http://www.allisons.org/ll/MML/">MML</A>].

%A C. S. Wallace
%T A Brief History of MML (Seminar)
%M NOV
%D 2003
%K CSW, CSWallace, c2003, c200x, c20xx, seminar, minimum message length, MML,
   minimum description length, MDL, NML, development, introduction, beginnings,
   origin, history, survey, AIC, BIC, complexity, comparison, compare, L3ref,
   Wal03, zMML, Bayesian, inductive inference, L4ref, II, AI, stats, statistics,
   CSci, machine learning, csse, monash university, algorithmic,
   research, computer science, Comp Sci, CompSci, CSci, talk, lecture, seminars
%X [<A HREF="http://www.allisons.org/ll/MML/20031120e/">transcribed</A>]['04].
   [Also search for: MML] and [also search for: CSWallace seminar].

%A C. S. Wallace
%A D. L. Dowe
%T MML clustering of multi-state, Poisson, von Mises circular and
   Gaussian distributions
%J Statistics and Computing
%V 10
%N 1
%P 73-83
%M JAN
%D 2000
%K jrnl, SNOB, minimum message length MML, inductive inference II, description,
   Normal, vonMises, classification, cluster, mixture, modelling, MDL,
   multistate, stats, AI, learning, data mining, CSW, CSWallace, DLD, zz1299,
   c2000, c200x, c20xx
%X Hint, [also search for: Snob]  or  more generally [also search for: MML];
   see [<A HREF="http://www.allisons.org/ll/MML/">MML page</A>].
   von Mises, probability density:
     f(x | mu, kappa) = (1/(2.pi.I0(kappa))).exp(kappa.cos(x-mu))
     where I0(kappa) is a normalisation constant.
   Its Fisher information
     F(mu,kappa) = N**2.kappa.A(kappa).{1-A(kappa)/kappa-(A(kappa))**2}
     where I1(kappa) = d I0(kappa)/d kappa
     and   A(kappa) = I1(kappa)/I0(kappa).
   -- [<A HREF="http://dx.doi.org/10.1023/A:1008992619036">doi:10.1023/A:1008992619036</A>][11/'04] (subscr'n?).
   Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW's publications</A>].

%A C. S. Wallace
%T Minimum description length
%B The MIT Encyclopedia of the Cognitive Science
%E R. A. Wilson
%E F. C. Keil
%I MITpress
%P 550-551
%D 1999
%K CSW, CSWallace, review, MDL, AI, II, c1999, c199x, c19xx
%X (from list prep by G.G. 12/'04.)  In, isbn:026273124X.
   (Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].)

%A C. S. Wallace
%A D. L. Dowe
%T Minimum message length and Kolmogorov complexity
%J COMPJ
%V 42
%N 4
%P 270-283
%D 1999
%K CSW, CSWallace, DLD, COMPJ, jrnl, MML, SMML, minimum description length,
   algorithmic complexity, AC, Kolmogorov complexity, KC, probability,
   prediction, predict, predicting, MDL, statistical inference,
   inductive inference, II, induction, AI, UTM, machine learning, Rissanen,
   Snob, CS, CSci, Monash, zz1099, c1999, c199x, c19xx
%X In a special issue of the Computer Journal (so is Solomonoff)
   on minimum length methods for machine learning.  [36 refs]
     1. Intro: (a) Kolmogorov '65, Lof, Chaitin '66, ...
               (b) Solomonoff '64, ...
               (c) Wallace & Boulton '68, ...; Rissanen '78, ...;
     2. Tech Details;  3. Data and Hypotheses;  4. 2-Part Encoding;
     5. Expected and Actual Lengths;  6. Choosing a Decoder;
     7. Terms of `order one';  8. Prediction and Induction;  9. Conclusion
   See also discussion
     by  `D a w i d' p323-326,   by `R i s s a n e n' p327-329,
     by  `C l a r k e' p338-339, by `S h e n' p340-342',
     and `Refinements of MDL and MML coding' by `W and D' p330-337,
     a reply to J. J. Rissanen's article, pp330-337,
     and a rejoinder by R p343-344, and one by `W and D' p345-347.
   -- [<A HREF="http://dx.doi.org/10.1093/comjnl/42.4.270">doi:10.1093/comjnl/42.4.270</A>]['05].
   Also see the [<A HREF="http://www.allisons.org/ll/MML/">MML page</A>].

%A C. S. Wallace
%T On the selection of the order of a polynomial model
%J Proc. of the 14th Biennial Australian Statistical Conf.
%E W. Robb
%W Queensland
%P 145
%M JUL
%D 1998
%O Royal Holloway College, London, '97
%K conf, minimum message length, MML, CSW, CSWallace, c1998, c199x, c19xx, MDL,
   stats, regression, description, polynomials, polynomial fitting, degree,
   function approximation, Vapnik-Chervonenkis VC dimension
%X "A report by Cherkassy, Mulier & Vapnik has compared the performance of
   several methods of selecting the order of a polynomial approximation to a
   fn t(x) given only the values of t(x) for some set of x-values, where the
   given values are corrupted by i.i.d. Gaussian noise of unknown variance. They
   compare various "classical" methods with a new method based on the
   Vapnik-Chervonenkis (VC) dimension, & conclude that the latter gives the most
   reliable prediction of the value of t(x) at unseen values of x. This work
   replicates their investigation & extends it by including a method based on
   Minimum Message Length (MML) estimation. The results largely confirm the
   previous results, but show that MML is generally superior to the VC method in
   terms of average squared prediction error."  (unref conf paper)
   -- [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/Polynomial/">more</A>].
   Also see sec.6.7.2, pp.272-275 in: [also search for: CSWallace MML book].

%A C. S. Wallace
%T A Monash FIT Graduation Address
%M APR
%D 1998
%K CSWallace, CSW, Monash, Faculty, FIT, ceremony, c1998, c199x, c19xx,
   altruism, computing, commercial, ethics, tragedy of the commons,
   open source, openSource, software
%X "First of all, let me congratulate you on your success. [...]
   I would like to offer you one final notion before you go, whether grain or
   chaff is for you to decide. There is a story told in economics called
   "The Tragedy of the Commons." [...]
   Considering the commercial importance and scale of the computing industry,
   a surprising number of the seminal developments in its technology have come
   from non-commercial sources. [...]"
   -- [<A HREF="/~gopal/cswgraduation.html">29 April 1998</A>],
   (Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].)

%A C. S. Wallace
%T MML inference of predictive trees, graphs and nets
%J Proc. Applied Decision Technologies 1995 Conf.,
   Stream 1: Computational Learning and Probabilistic Reasoning
%W London
%P 101-115
%D 1995
%K conf, MML, minimum message length, description, inductive inference, MDL,
   II, AI, predict, prediction, expert systems, decision, classification,
   DTree, tree, net, graph, c1995, c199x, c19xx, graph, net, CSW, CSWallace
%X via GF. Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T How to do Inductive Inference using Information Theory (Seminar)
%D 1993
%K Wal93,zzMML, seminar, c1993, c199x, c19xx, CSW, CSWallace, Monash, MML, MDL,
   AIC, AI, II, minimum message length, description, explanation, mmld, stats,
   statistics, Bayesian inference, universal TM, UTM, machine learning,
   total ignorance, prior, priors, computer science, Comp Sci, CompSci,
   talk, lecture, seminars
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/1993-08-10/">transcribed</A>]['11].
   [Also search for: CSWallace seminar].

%A J. Oliver
%A D. L. Dowe
%A C. S. Wallace
%T Inferring Decision Graphs Using the Minimum Message Length Principle
%J Proc. of the 1992 Aust. Joint Conf. on Artificial Intelligence
%W Hobart, Tasmania
%P 361-367
%M SEP
%D 1992
%K conf, Monash, CSW, AJCAI AJCAI92, minimum message length MML, encoding,
   description, MDL, inductive inference, II,
   decision, graph, tree, CSW, CSWallace, c1992, c199x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].
   [Also search for: Dgraph decision].

%A C. E. Chew
%A C. S. Wallace
%T An Inter-Bus Connection for a Capability Based Multiprocessor
%J Proc. of 14th Australian Computer Science Conf.
%W Sydney,
%O Australian Computer Science Communications, Vol.13, No.1, 30/1-10,
%D 1991
%K Monash, CSW, ACSC, ACSC14, ACSC91, multi processor, bus, CSW, CSWallace,
   computer architecture, hardware, c1991, c199x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%A R. D.  Pose
%T Charging in a Secure Environment
%E J. Rosenberg
%E J. L. Keedy
%J Security and Persistance
%W Bremen
%D 1990
%I SpringerVerlag
%P 85-97
%K CSW, CSWallace, Monash, operating system, OS, charge, money, persistence,
   conf, capability, persistant, multiprocessor, computer, multi, security,
   c1990, c199x, c19xx
%X "The Monash Multiprocessor Architecture incorporates a monetary system at the
   lowest kernel level, integrated with a password capability scheme. Although
   the capability scheme is quite flexible, providing support for non-hierarchic
   security & access policies, abstract type management & information
   confinement, we show that it is possible for service providers to command
   use-based fees for service. The fee charging protocols must be designed with
   some care to avoid breaching required information confinement constraints
   when user & provider are mutually suspicious, but need not be v.complicated"
   -- [<A HREF="https://doi.org/10.1007/978-1-4471-3178-6_6">doi:10.1007/978-1-4471-3178-6_6</A>]['17].
   Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A E. D. Cathro
%A C. S. Wallace
%A M. S. Anderson
%T An I/O Subsystem for a Multiprocessor
%J Proc. of the 12th Australian Computer Science Conf.
%V 11
%N 1
%P 275-285
%M FEB
%D 1989
%K CSW, CSWallace, Monash, ACSC ACSC12 ACSC89, multi processor, input, output,
   computer architecture, IO, c1989, c198x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%A et al (Multi Group)
%T Using a Unix System as a Multiprocessor Frontend
%J AUUGN
%V 7
%N 2-3
%P 15-21
%D 1988 ???
%K conf, CSW, CSWallace, Monash, multi processor, c1988, c198x, c19xx,
   operating system, OS
%X date ~ c198x. [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A M. S. Anderson
%A C. S. Wallace
%T Some Comments on the Implementation of Capabilities
%J The Australian Computer Journal
%V 20
%N 3
%P 122-133
%M AUG
%D 1988
%K jrnl, CSW, CSWallace, ACJ, Monash, capability, operating system OS,
   c1988, c198x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A R. D. Pose
%A M. S. Anderson
%A C. S. Wallace
%T Implementation of a Tightly-Coupled Multiprocessor
%J Proc. ACSC-10
%V 9
%N 1
%P 330-342
%M FEB
%D 1987
%K conf, CSW, CSWallace, Monash, ACSC ACSC10 ACSC87, multi processor,
   computer architecture, c1987, c198x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T An improved program for classification
%J Proc. of ACSC-9
%V 8
%N 1
%P 357-366
%M FEB
%D 1986
%O C. S. Wallace, "An improved program for classification",
      Monash U., Comp. Sci., TR47, 1984
%K CSW, conf, Monash, ACSC, ACSC9, ACSC86, class, classify, classification,
   coding, unsupervised, mixture model, modelling, cluster, clustering, SNOB,
   bit borrowing, hard, fractional, soft assignment, II, minimum message length,
   MML, MDL, bits back, borrow, description, Hinton, Camp, COLT, encoding,
   Snob2, dld, SMEM, EM algorithm, c1986, c198x, c19xx, TR47,c1984, CSWallace
%X The first description of the bit-borrowing coding technique later called
   bits-back by G. E. Hinton & D. van Camp, COLT '93, (COLT93) 1993.
   (Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].)
   [Also search for: Wallace ICCI].

%A R. D. Pose
%A M. S. Anderson
%A C. S. Wallace
%T Aspects of a Multiprocessor Architecture
%J Proc. Workshop on Future Directions in Computer Architecture and Software
%W Charlston
%P 293-295
%M MAY
%D 1986
%K CSW, Monash, multi processor, CSWallace, c1986, c198x, c19xx, computer,
   wShop
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A Wallace C.S,
%A Koch D.,
%T TTL - Compatible Multiport Bus
%J Computer Systems Science and Engineering
%V 1
%N 1
%P 47-52
%M OCT
%D 1985
%K CSW, CSWallace, Monash, multi processor, c1985, c198x, c19xx,
   computer architecture
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A G. K. Gupta
%A C. S. Wallace
%T A New Step-Size Changing Technique for Multistep Methods
%J Mathematics of Computation
%V 33
%N 145
%P 125-138
%M JAN
%D 1979
%K jrnl, CSW, CSWallace, Monash, multi, step, size, c1979, c197x, c19xx,
   numerical analysis, methods
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T Memory and Addressing Extensions to a HP2100A
%J Proc. 8th Australian Computer Conf.
%V 4
%P 1796-1811
%M AUG-SEP
%D 1978
%K CSW, CSWallace, Monash, c1978, c197x, c19xx, computer architecture
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T Computing Research in Australia
%J The Australian Computer Journal
%V 9
%N 1
%P 21-24
%M MAR
%D 1977
%K jrnl, CSW, CSWallace, Monash, ACJ, c1977, c197x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A R. A. Hagan
%A C. S. Wallace
%T A Virtual Memory System for the HP2100A
%J Computer Architecture News
%V 6
%N 5
%P 5-13
%D 1977
%K jrnl, CSW, computer architecture, CSWallace, Monash, c1977, c197x, c19xx,
   virtual memory, paged, paging
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T Transformed rejection generators for Gamma and Normal pseudorandom variables
%J The Australian Computer Journal
%V 8
%N 3
%P 103-105
%M NOV
%D 1976
%K jrnl, CSW, CSWallace, Monash, pseudo random number generator, RNG, PRNG,
   gamma, normal, Gaussian, probability distributions, c1976, c197x, c19xx, ACJ
%X (Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].)

%A C. S. Wallace
%A D. Harper
%A R. A. Hagan
%T A Discrete System Simulation Package for a Mini Computer
%J ACM Simuletter
%V 7
%N 1
%P 9-13
%D 1976
%K CSW, CSWallace, Monash, c1976, c197x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A D. M. Boulton
%A C. S. Wallace
%T An information measure for single-link classification
%J COMPJ
%V 18
%N 3
%P 236-238
%M AUG
%D 1975
%K jrnl, CSW, CSWallace, Monash, class, classify, classification, cluster,
   clustering, stats, SNOB, minimum message length, MML, encoding, description,
   minimum spanning tree, trees, linkage, MST, MDL, c1975, c197x, c19xx
%X "The information measure is an objective measure of the quality of a
   classification and results from an information transmission view of the
   classification problem. So far information measures have only been derived
   for the case where an explicit assumption is made about the form of the
   distribution of attribute values within a class. One important method which
   involves no such explicit assumption is single link. In this paper we derive
   a new information measure which is optimised by classifications produced by
   the single link method. By investigating the properties of this information
   measure we are able to gain more insight into the single link method and
   also determine the type of problem to which it best applies."
   -- [<A HREF="http://dx.doi.org/10.1093/comjnl/18.3.236">doi:10.1093/comjnl/18.3.236</A>]['05].
   Also see [<A HREF="http://www.allisons.org/ll/MML/">MML</A>].
   [Also search for: single link]  and  [also search for: MML].

%A G. K. Gupta
%A C. S. Wallace
%T Some New Multistep Methods for Solving Ordinary Differential Equations
%J Proc. of ANZAAS Conf.
%D 1973
%O Math. of Comp., 29, 130, pp.489-500, April 1975
%K CSW, CSWallace, Monash, multi step, ODE, ODEs, c1973, c197x, c19xx,
   numerical methods, analysis
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%A D. M. Boulton
%T An invariant Bayes method for point estimation
%J Classification Society Bulletin
%V 3
%N 3
%P 11-34
%D 1975
%K jrnl, CSW, WB75, Monash, Bayes, Bayesian, statistics, minimum message length,
   MML, SMML, encoding, description, MDL, stats, c1975, c197x, c19xx, CSWallace,
   estimate, model, II, inference
%X "We consider the problem of forming a point estimate of an unknown
   parameter, based on the result of an experiment, & using, via Bayes' theorem,
   prior information about the parameter. ..."
   Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%A G. K. Gupta
%T General Linear Multistep Methods to Solve Ordinary Differential Equations
%J The Australian Computer Journal
%V 5
%N 2
%P 62-69
%M MAY
%D 1973
%K jrnl, CSW, CSWallace, Monash, multi step, ODE, ODEs, c1973, c197x, c19xx,
   ACJ, numerical methods, analysis
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A D. M. Boulton
%A C. S. Wallace
%T A comparison between information measure classification
%J Proc. of ANZAAS Congress
%W Perth
%M AUG
%D 1973
%K CSW, Monash, class, classify, classification, cluster, clustering, SNOB,
   mixture model, modelling, minimum message length, MML, encoding, MDL,
   CSWallace, c1973, c197x, c19xx, conf
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A D. M. Boulton
%A C. S. Wallace
%T Occupancy of a rectangular array
%J COMPJ
%V 16
%N 1
%P 57-63
%D 1973
%K jrnl, compj, chi squre, squared, association, rectangle, arrays, Bell,
   integer, CSW, Monash, c1973, c197x, c19xx, CSWallace
%X "The problem addressed is: the occupancy, by non-negative integers, of a
   rectangular array of given size when the row sums and column sums are all
   fixed. In general, a single expression for the number of array element
   combinations which covers all combinations of row and column sums cannot
   be found. Thus two algorithms which systematically count all element
   combinations are presented, the first applying to arrays with only two
   columns and the second to arrays with any number of rows and columns.
   The two-column algorithm is then used to derive expressions for the number
   of array element combinations of arrays having three rows and two columns.
   Five different expressions are required to cover all combinations of row and
   column totals."
   [<A HREF="http://dx.doi.org/10.1093/comjnl/16.1.57">doi:10.1093/comjnl/16.1.57</A>]['05]
   Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T Simulation of a two-dimensional gas
%J Proc. of ANZAAS Conf.
%W Perth
%P 19
%M AUG
%D 1973
%K CSW, CSWallace, Monash, 2D, two dimensional, gas, c1973, c197x, c19xx,
   thermodynamics, entropy, physics, theory, law
%X (Also see [<A HREF="http://www.allisons.org/ll/MML/Entropy/Gas/">gas</A>]['01] simulation.)

%A P. S. Roberts
%A C. S. Wallace
%T A Microprogrammed Lexical Processor
%J Information Processing 71
%P 577-581
%D 1972
%K CSW, CSWallace, Monash, c1972, c197x, c19xx, computer hardware, firmware,
   conf, microprogramming, architecture
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A A. Y. Montgomery
%A C. S. Wallace
%T Evaluation and Design of Random Access Files
%J Proc. 5th Australian Computer Conf.
%W Brisbane
%P 142-150
%M MAY
%D 1972
%K CSW, CSWallace, Monash, ACC, ACC5, ACC72, c1972, c197x, c19xx, file system
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T Simulation Studies in File Structures
%J A.C.S Victorian Branch Professional Development Workshop
%P 16
%M FEB
%D 1971
%K wShop, CSW, CSWallace, Monash, c1971, c197x, c19xx, file system, structure
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A D. M. Boulton
%A C. S. Wallace
%A A program for numerical classification
%J COMPJ
%V 13
%N 1
%P 63-69
%M FEB
%D 1970
%K jrnl, COMPJ, c1970, c197x, c19xx, CSWallace, CSW, Monash, class, classify,
   classification, cluster, clustering, mixture modelling, model, stats, SNOB,
   BW60, minimum message length, II, MML, encoding, MDL, bit, nit, nat, ban,
   information theory
%X "In a previous paper (W & B '68) the information measure was derived. It is
   designed to measure the objective goodness of a nonhierarchical taxonomic
   classification & can be used to choose the best of a number of different
   classifications of the one data set. The information measure can also form
   the basis of a classification alg. which searches directly for that
   classification with the best information measure. ... such a classification
   alg. is described together with an ALGOL program called Snob ..."
   -- [<A HREF="http://dx.doi.org/10.1093/comjnl/13.1.63">doi:10.1093/comjnl/13.1.63</A>]['05];
   see [<A HREF="http://www.allisons.org/ll/MML/">MML</A>],
   [also search for: MML].

%A C. S. Wallace
%A A. M. Jessup
%A A Simple Graphic Input/Output Device
%J The Australian Computer Journal
%V 2
%N 1
%P 39-40
%M FEB
%D 1970
%K jrnl, CSW, CSWallace, Monash, ACJ, input output, c1970, c197x, c19xx,
   computer architecture, hardware, graphics
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A G. M. Clarke
%A C. S. Wallace
%T Analysis of Nasal Support
%J Archives of Otolaryngology
%V 92
%P 118-129
%M AUG
%D 1970
%K jrnl, CSW, CSWallace, Monash, nose, nasal, medical, c1970, c197x, c19xx
%X (!).  [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T A Digital Logic Simulator for Teaching Purposes
%J Proc. of the 4th Australian Computer Conf.
%P 215-219
%D 1969
%K CSW, CSWallace, Monash, ACC, ACC4, ACC69, c1969, c196x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A T. J. McQuade
%A D. Race
%A C. S. Wallace
%T Storage and Retrieval of Medical Information
%J Proc. of the 4th Australian Computer Conf.
%P 539-544
%D 1969
%K CSW, CSWallace, Monash, ACC ACC4 ACC69, medical informatics,
   c1969, c196x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T Control Systems
%J The Fourth Australian Computer Conf.
%V 2
%N 4
%P ?-?
%M AUG
%D 1969
%K CSW, CSWallace, Monash, c1969, c196x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A A. M. Jessup
%A C. S. Wallace
%T A Cheap Graphic Input Device
%J Aust. Comp. J.
%V 1
%P 2
%M MAY
%D 1968
%K jrnl, CSW, CSWallace, Monash, ACJ, c1968, c196x, c19xx, computer graphics
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%A B. G. Rowswell
%T Competition for memory access in the KDF9
%J COMPJ
%V 10
%N 1
%P 64-68
%M MAY
%D 1967
%K COMPJ, jrnl, CSW, CSWallace, KDF9, KDF 9, operating system, OS, usage, use,
   round robin, first come first served, priority, c1967, c196x, c19xx,
   computer architecture, hardware
%X "The I/O control of a computer may adopt various strategies for serving
   competing requests from peripheral channels for access to a single core
   memory. The strategy adopted places some limit on the peripheral
   configurations which may be simultaneously active. An "ideal" strategy is
   exhibited which imposes no constraint other than that implied by the finite
   speed of the core memory, but it is expensive to implement. Limits on
   configurations are derived for the ideal, first come-first served,
   round-robin & priority systems. It is shown that the first come-first served
   & round-robin systems have little or no advantage over a random choice. The
   KDF9 at the U. of Sydney, which was delivered with a first come-first
   served system, has been modified to incorporate a priority system. The
   new system uses 48 fewer circuit boards, but allows the attachment of
   high-rate channels which otherwise could not be accommodated."
   -- [<A HREF="http://dx.doi.org/10.1093/comjnl/10.1.64">doi:10.1093/comjnl/10.1.64</A>]['05],
   and [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].
   [Also search for: KDF9].

%A C. S. Wallace
%A O. Longe
%T Reading gapless tapes
%J IEEE Trans. Elec. Comp.
%V EC-16
%N 4
%P 517-518
%M AUG
%D 1967
%K jrnl, magnetic tape, computer hardware, CSW, CSWallace, c1967, c196x, c19xx,
   EC16, KDF 9, KDF9
%X "Field-recorded data are often most conveniently recorded on magnetic tapes
   without interblock gaps. Gapless tapes cannot be read by certain types of
   computers. A simple modification to a KDF 9 computer enables it to transcribe
   from gapless to conventional format as a time-shared monitor operation. The
   counter which increases peripheral transfer addresses and detects the
   completion of a block is time-shared by all channels on the KDF 9. It was
   modified so that for any channel which, in requesting a core memory access,
   indicates that it is engaged in a gapless read" carries out of the 64's digit
   of the counter are suppressed, and any change of the 64's digit causes an
   interrupt to the permanently resident monitor program (Director). Thus, if
   an input transfer is set up on a channel with initial core memory address an
   exact multiple of 128, and the channel is made to signal that it is reading
   gapless tape, the reading will proceed indefinitely, repeatedly cycling over
   the same 128 word core area, and causing an entry to Director after every 64
   words."
   -- [<A HREF="http://dx.doi.org/10.1109/PGEC.1967.264681">doi:10.1109/PGEC.1967.264681</A>]['11].
   [Also search for: KDF9].
   (Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].)

%A C. S. Wallace
%T Digital Computers
%P ?-?
%B Atoms to Andromeda
%E S. T. Butler
%E H. Messel
%I ShakespeareHead, Sydney
%D 1966
%K CSW, CSWallace, c1966, c196x, c19xx, c1966, c196x, c19xx, chapter, computer
%X "selected lectures on theoretical physics, high-energy nuclear and
   cosmic ray research, plasma and thermonuclear physics, astronomy,
   astrophysics and electronic computing."   Also see
   [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A Boden
%A Branagan
%A Davies
%A Gould
%A Graham
%A Kelly
%A Mercer
%A Ross
%A Strahan
%A Wallace
%T Advancing with Science
%I Science Press, Sydney
%D 1966
%K CSW, CSWallace, c1966, c196x, c19xx, chapter
%X (This is a textbook for secondary schools, of which csw wrote one chapter
   and contributed substantially to four others.)
   [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A J. M. Bennett
%A C. S. Wallace
%A J. W. Winings,
%T Software and Hardware Contributions to Efficient Operating on a Limited
   Budget
%J Proc. 3rd Australian Comp. Conf.
%W Canberra
%P ?-?
%D 1966
%K conf, CSW, CSWallace, ACC, ACC3, ACC66, c1966, c196x, c19xx, computing
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A I. C. A. Martin
%A C. S. Wallace
%T Impedance Change Frequency of Diluted Ram Semen recorded on a Digital Scaler
%J J. Reprod. Fertil.
%V 10
%P 425-437
%D 1965
%K jrnl, CSW, CSWallace, biology, biol, c1965, c196x, c19xx, reproductive,
   reproduction, animal fertility
%X The title says it all,
   [<A HREF="http://www.reproduction-online.org/content/vol10/issue3/index.shtml">jrnl_contents</A>]['04].
   Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T A suggestion for a fast multiplier
%J IEEE Trans. on Electronic Comp.
%V EC-13
%N 1
%P 14-17
%D 1964
%O Univ' Illinois, TR UIUCDCS-R-63-133, '63
%K CSW, CSWallace, Wallace Tree multiplier, multiply, multiplication, CPU, ALU,
   computer hardware, architecture, circuit, arithmetic, algorithm, square root,
   W64, c1964, c196x, c19xx, jrnl
%X "It is suggested that the economics of present large-scale sci. computers
   could benefit from a greater investment in h/w to mechanize multiplication
   and division than is now common. As a move in this direction, a design is
   developed for a multiplier which generates the product of two numbers using
   purely combinational logic, i.e., in one gating step. Using straightforward
   diode-transistor logic, it appears presently possible to obtain products in
   under 1, mu-sec, and quotients in 3 mu-sec. A rapid square-root process is
   also outlined. Approximate component counts are given for the proposed
   design, and it is found that the cost of the unit would be about 10 per cent
   of the cost of a modern large-scale computer."
   -- [<A HREF="https://doi.org/10.1109/PGEC.1964.263830">doi:10.1109/PGEC.1964.263830</A>]['17].
   Reduces the problem of multiplying two n-bit numbers to one of
   summing two O(n)-bit numbers.
   Also see
   [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>], and
   w.t.@[<A HREF="http://en.wikipedia.org/wiki/Wallace_tree">wikip</A>]['11].
   [Also search for: Wallace multiplier].

%A C. S. Wallace
%T Correlated round-off errors in digital integrating differential analyzers
%J COMPJ
%V 7
%N 2
%P 131-134
%M JUL
%D 1964
%K jrnl, compj, architecture, hardware, error, analyzer, c1964, c196x, c19xx,
   CSW, CSWallace, computer, hadware, integration
%X "The integration method used in digital differential analyzers suffers from
   both truncation and round-off errors. It is shown that, if trapezoidal
   corrections are employed, the latter dominates. The round-off errors in
   successive steps of the integration are shown to be correlated whenever the
   function being integrated has a slope which passes through a rational
   fraction of small denominator. However, an analysis is presented to show
   that in general the correlation does not greatly affect the total round-off
   error. Some special cases are shown to suffer from anomalously high
   round-off error."
   [<A HREF="http://dx.doi.org/10.1093/comjnl/7.2.131">doi:10.1093/comjnl/7.2.131</A>]['05],
   [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A J. Malos
%A D. D. Millar
%A C. S. Wallace
%T Cerenkov radiation from E.A.S.
%J J. Phys. Soc. Japan
%V 17 Supplement A - III
%P ?-?
%D 1962
%K jrnl, physics, EAS, CSW, CSWallace, c1962, c196x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T Comparison between the response of Geiger and scintillation counters to
   the air shower flux
%J Proc. of Moscow Cosmic Ray Conf.
%V II
%P 316
%D 1960
%K conf, rays, CSW, CSWallace, physics, c1960, c196x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T Counter Experiments on Extensive Cosmic Ray Air Showers
%I U. Sydney
%D 1960
%K PhD Thesis, physics, CSW, CSWallace, c1960, c196x, c19xx, rays, shower

%A C. S. Wallace
%T The determination of the radial distribution of electrons, and the size
   spectrum of extensive air showers
%J Proc. of A.A.E.C. Symp.
%W Sydney
%P ?-?
%D 1958
%K conf, AAEC, physics, CSW, CSWallace, c1958, c195x, c19xx,
   cosmic ray, shower
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A M. H. Brennan
%A J. A. Lehane
%A J. Malos
%A D. D. Millar
%A C. S. Wallace
%A M. M. Winn
%T The Sydney air shower experiment
%J N.2 del Supplemento al Vol. 8, Serie X del Nuovo Cimento
%P 653-661
%D 1958 ?
%K physics, cosmic ray, rays, shower, CSW, CSWallace, c1958, c195x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A M. H. Brennan
%A J. Malos
%A D. D.  Millar
%A C. S. Wallace
%T Cerenkov light from air showers
%J N2 del Supplemento al vol.8, Serie X, del Nuovo Cimento
%P 662-664
%D 1958 ?
%K physics, cosmic ray, rays, shower, CSW, CSWallace, c1958, c195x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%A M. M. Winn
%A K. K. Ogilvie
%T Dependence of the nucleonic component of cosmic ray air showers on radius
%J Nature
%N 182
%N 4650
%P 1653-1654
%M DEC
%D 1958
%K jrnl, physics, c1958, c195x, c19xx, CSW, CSWallace, cosmic ray, rays,
   shower
%X [<A HREF="http://dx.doi.org/10.1038/1821653a0">doi:10.1038/1821653a0</A>]['07],
   [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A M. H. Brennan
%A D. D. Millar
%A C. S. Wallace
%T Air showers of size greater than 10**5 particles - (1) core location and
   shower size determination
%J Nature
%V 182
%P 905-911
%M OCT
%D 1958
%K jrnl, physics, c1958, c195x, c19xx, CSW, CSWallace, cosmic ray, rays,
   shower
%X [<A HREF="http://dx.doi.org/10.1038/182905a0">doi:10.1038/182905a0</A>]['07],
   [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A M. H. Brennan
%A J. Malos
%A D. D. Millar
%A M. H. Rathgeber
%A C. S.  Wallace
%T Air showers of size greater than 10**5 particles - (2) Cerenkov radiation
   accompanying the showers
%J Nature
%V 182
%P 973-977
%M OCT
%D 1958
%K jrnl, physics, cosmic ray, rays, CSW, CSWallace, c1958, c195x, c19xx
%X [<A HREF="http://dx.doi.org/10.1038/182973a0">doi:10.1038/182973a0</A>]['07],
   [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A M. H. Brennan
%A D. D. Millar
%A M. H. Rathgebar
%A C. S. Wallace
%T Air showers of size greater than 10**5 particles - (3) comparison
   between the response of Geiger and scintillation counters
%J Nature
%V 182
%M OCT
%D 1958
%P 1053-1054
%K jrnl, physics, CSW, CSWallace, c1958, c195x, c19xx
%X [<A HREF="http://dx.doi.org/10.1038/1821053a0">doi:10.1038/1821053a0</A>]['07],
   [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A B. J. O'Brien
%A C. S. Wallace
%T Ettingshausen effect and thermomagnetic cooling
%J J. of Applied Physics
%V 29
%P 1010-1012
%M JUL
%D 1958
%K jrnl, physics, efects, CSW, CSWallace, c1958, c195x, c19xx, thermo-magnetic
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%A M. H. Brennan
%T The automatic digital recording of information from cosmic ray air showers
%J Proc. of Salisbury Conf. on Computing, W.R.E., Salisbury, S.A.
%P ?-?
%M JUN
%D 1957
%K conf, physics, hardware, cosmic rays, CSW, CSWallace, c1957, c195x, c19xx,
   shower, ray
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A B. J. O'Brien
%A C. S. Wallace
%A K. Landecker
%T The cascading of Peltier-couples for thermo-electric cooling
%J J. of Applied Physics
%V 27
%P 820-823
%M JUL
%D 1956
%K jrnl, physics, thermoelectric, CSW, CSWallace, c1956, c195x, c19xx, couple,
   Peltier effect, refrigeration
%X <!-- from dept record --> [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A C. S. Wallace
%T Suggested design for a very fast multiplier
%R UIUCDCS-R-63-133
%D 1963
%I Dept Computer Science, University of Illinois at Urbana-Champaign
%P 25
%O IEEE Trans. Comput, V 13, N ?, p ??-??, '64
%K CSW, CSWallace, hardware, fast, multiplier, multiplication, Wallace,
   multiply, computer, algorithm, arithmetic, integer, tree, trees,
   TR, TR 63 133, TR133, c1963, c196x, c19xx
%X [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>],
   [also search for: Wallace multiplier].

%A C. S. Wallace
%T A long-period pseudo-random generator
%R 89/123,
%I Dept. Comp. Sci., Monash Uni.
%M FEB
%D 1989
%K TR 123, TR123, CSW, CSWallace, PRNG, RNG, number, numbers,
   c1989, c198x, c19xx
%X (Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].)

%A C. S. Wallace
%T Fast pseudo-random generators for Normal and Exponential variates
%J ACM Trans. Math. Software
%V 22
%N 1
%P 119-127
%M MAR
%D 1996
%O TR 94/197, May 1994, Dept. Comp. Sci., Monash Uni.
%K jrnl, TOMS, c1996, c199x, c19xx, CSW, CSWallace, Monash, RNG, PRNG,
   pseudo random number generator, algorithm, numbers, normal, probability,
   distribution, variate, Gaussian, distribution, TR 94 197, TR197, generate
%X "Fast algorithms for generating pseudorandom numbers from the unit-normal and
   unit-exponential distributions are described. The methods are unusual in that
   they do not rely on a source of uniform random numbers, but generate the
   target distributions directly by using their maximal-entropy properties. The
   algorithms are fast. The normal generator is faster than the commonly used
   Unix library uniform generator 'random' when the latter is used to yield real
   values. Their statistical properties seem satisfactory, but only a limited
   suite of tests has been conducted. They are written in C and as written
   assume 32-bit integer arithmetic. The code is publicly available as C source
   and can easily be adopted for longer word lengths and/or vector processing."
   -- [<A HREF="http://dx.doi.org/10.1145/225545.225554">doi:10.1145/225545.225554</A>]['11];
   also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW publications</A>],
   [<A HREF="ftp://ftp.csse.monash.edu.au/pub/csw">CSW ftp</A>]['96].

%A M. Castro
%A G. Pringle
%A C. S. Wallace
%T The Walnut kernel: program implementation under the Walnut kernel
%I Dept. Comp. Sci., Monash Uni., Australia 3168
%R 95/230
%M AUG
%D 1995
%K TR 230, TR230, csw, operating system, OS, secure, security, capability,
   object, walnut, kernel, CSci, zz0995, c1995, c199x, c19xx, CSW, CSWallace
%X Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].

%A D. L. Dowe
%A J. J. Oliver
%A R. A. Baxter
%A C. S. Wallace
%T Bayesian estimation of the Von Mises concentration parameter
%J Proc. 15th Int. Workshop on Maximum Entropy and Bayesian Methods, Santa Fe
%P 51-60
%M JUL
%D 1995
%O TR 95/236, Dept. Comp. Sci., Monash Uni., Sept 1995
%K TR 236, TR236, vonMises, vM, circle, circular, probability distribution,
   inductive inference, II, MML, CSW, CSWallace, DLD, jono, MaxEnt, Max Ent,
   c1995, c199x, c19xx, zz0995
%X "... examine a variety of Bayesian estimation techniques by examining the
   posterior distn in both polar & Cartesian co-ordinates. We compare the MML
   est. with these fellow B.techniques, & a range of Classical ests.. We find
   that the B. ests. outperform the Classical ests.."
   -- [<A HREF="https://doi.org/10.1007/978-94-011-5430-7_6">doi:10.1007/978-94-011-5430-7_6</A>]['19].
   (Later MML research: [<A HREF="http://www.allisons.org/ll/MML/">MML</A>].)

%A C. S. Wallace
%A K. Korb
%A H. Dai
%T Causal discovery via MML
%J Proc. 13th Int. Conf. on Machine Learning
%E L. Saitta
%I MorganKaufmann
%P 516-524
%D 1996
%O TR 96/254, Dept. Comp. Sci., Monash Uni., Australia 3168,
   Feb 1996
%K conf, ICML, ICML13, causal net, nets, cause and effect, inductive inference,
   II, description, TR 254, TR254, minimum message length, MML, MDL, KBKorb,
   CSW, CSWallace, model, models, zz0296, c1996, c199x, c19xx
%X Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].
   [Also search for: II Bnets].

%A H. Dai
%A K. Korb
%A C. Wallace
%T A study of causal discovery with weak links and small samples
%I Dept. Comp. Sci., Monash Uni., Australia 3168
%R 96/298
%M DEC
%D 1996
%O Intelligent Information Systems, pp.27-30, 1996,
   Australian and New Zealand conf.
%K conf, TR 298, TR298, CSW, CSWallace, cause, effect, network, MML, MDL,
   KBKorb, machine learning, AI, inductive inference, II,
   zz1296, c1996, c199x, c19xx
%X "... examines the influence of sample size on the discovery of causal models.
   The experimental results illustrate the effect of larger sample sizes for
   reliably discovering causal models and the relevance of the strength of
   causal links and the complexity of the original causal model. They present
   indicative evidence of the superior robustness of MML methods to standard
   significance tests in the recovery of causal links. The comparative results
   show that the MML ..."
   (Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].)

%A C. S. Wallace
%A K. B. Korb
%T Learning linear causal models by MML sampling
%B Causal Models and Intelligent Data Management
%E A. Gammerman
%I SpringerVerlag
%P 89-111
%D 1999
%O TR 97/310, Dept. Comp. Sci., Monash Uni., Australia, June '97
%K TR310, TR 310, model, cause, effect, inductive inference, MML, MDL, zz0697,
   conf, CSW, CSWallace, AI, learning, CS, CSci, minimum message length, MCMC,
   KBKorb, CaMML, causal network, machine learning, AI, II, net, nets, network,
   TETRAD, Madigan, description, zz1098, c1999, c199x, c19xx
%X "We combine [MML] evaln of linear causal models with Monte Carlo sampling to
   produce a program that, given ordinary joint sample data, reports the
   posterior probs. of equiv. classes of causal models & their member models.
   We compare our program with TETRAD II [7.15] & the Bayesian MCMC program of
   Madigan et al. [7.11]. Our approach differs from that of M. et al. [7.11]
   particularly in not assigning equal prior probs. to equiv. classes of causal
   models & in merging models from distinct equiv. classes when the causal links
   are suff. weak that the sample data available could not be expected to
   distinguish between them (which we call 'small effect equivalence')."
   -- [<A HREF="http://dx.doi.org/10.1007/978-3-642-58648-4_7">doi:10.1007/978-3-642-58648-4_7</A>]['14] (1999).
   See [<A HREF="/~lloyd/tildeImages/People/Wallace/">CSW</A>].
   [Tech report #310 June 1997, conf '98? / book 1999.]

%A K. B. Korb
%A C. S. Wallace
%T In search of the philosopher's stone: Remarks on Humphreys and
   Freedman's critique of causal discovery
%J British Jrnl. for the Philosophy of Science
%P 543-553
%D 1999
%O TR 97/315, Mar 1997, Dept. Comp. Sci., Monash Univ., Australia 3168
%K TR 315, TR315, CSW, CSWallace, causal, cause and effect, CaMML, AI, zz0397,
   c1999, c199x, c19xx, KBKorb
%X See [<A HREF="/~lloyd/tildeImages/People/Wallace/">CSW</A>].

%A G. E. Farr
%A C. S. Wallace
%T The complexity of strict minimum message length inference
%J COMPJ
%V 45
%N 3
%P 285-292
%D 2002
%O TR 97/321, Dept. Comp. Sci., Monash Uni., Australia 3800, Aug 1997
%K CompJ, Jrnl, BCS, 1997, TR321, TR 321, algorithmic complexity, NP, NPC, NPH,
   theory, SMML, MML, MDL, AI, II, CSW, CSWallace, minimum message length, BIC,
   description, binomial, trinomial, multinomial, probability distribution,
   zz0897, zz0602, c2002, c200x, c20xx, algorithmic complexity, criterion,
   information theory, optimal, encoding
%X "Strict Minimum Message Length (SMML) inference is an information-theoretic
   criterion for inductive inference introduced by Wallace and Boulton and
   is known to possess several desirable statistical properties. In this paper
   we examine its computational complexity. We give an efficient algorithm for
   the binomial case and indeed for any SMML problem that is essentially
   one-dimensional in character. The problem in general is shown to be NP-hard.
   A heuristic is discussed which gives good results for binomial and trinomial
   SMML inference. The complexity of the trinomial case remains open and is
   worth further investigation."
   -- [<A HREF="http://dx.doi.org/10.1093/comjnl/45.3.285">doi:10.1093/comjnl/45.3.285</A>]['05]
   (Also see [<A HREF="http://www.allisons.org/ll/MML/SMML/Binomial/">SMML</A>].)

%A J. R. Neil
%A C. S. Wallace
%A K. B. Korb
%T Bayesian networks with non-interacting causes
%I School of Comp. Sci. & Software Engineering,
   Monash Uni., Australia 3168
%R 1999/28
%M SEP
%D 1999
%K TR28, TR 28, network, cause, causal, effect, model, AI, MML, MDL, zz0999,
   CSW, CSWallace, KBKorb, JRNeil, c1999, c199x, c19xx
%X (that's right, Sept '99.)
   (Also see [<A HREF="http://www.allisons.org/ll/Images/People/Wallace/">CSW</A>].)

%A C. Kopp
%A C. S. Wallace
%T TROPPO - A tropospheric propagation simulator
%I School of Computer Science and Software Engineering,
   Monash Uni., Australia 3800
%R 2004/161
%M NOV
%D 2004
%K TR 161, TR161, CSW, CSWallace, troposphere, data communications, comms,
   radio, zz1104
%X CK with the late [<A HREF="/~lloyd/tildeImages/People/Wallace/">CSW</A>].

—ip