Monash Univ Fellowships (2+ years postPhD)

Monash University

Monash University has established the prestigious Monash Fellowships Program to attract internationally competitive researchers from any discipline to undertake research of national and international significance and provide them with the best possible research environment.

The Fellowships will be in the form of a direct appointment following a Monash-conducted competitive process open to researchers from outside Monash University, from within and outside Australia.

Please note that applicants should normally have at least two (2) but not more than eight (8) years of post-doctoral research experience and [of course :-) ] demonstrate excellent research outcomes at international level

Assoc. Prof. David Dowe (www.csse.monash.edu.au/~dld , www.csse.monash.edu.au/~dld/David.Dowe.publications.html ) seeks applicants for one of these Monash University postdoctoral fellowships in any of a wide range of areas primarily pertaining to the Bayesian information-theoretic Minimum Message Length (MML) principle (Wallace & Boulton, Computer J, 1968; Wallace & Freeman, J Royal Stat Soc 1987).

Despite the beliefs of many that it is not possible to have a statistically invariant Bayesian method of point estimation, MML is just that. For a fair but pretty stark comparison between MML and Akaike's Information Criterion (AIC), see, e.g., http://dx.doi.org/10.1093/bjps/axm033 (Brit. J. Phil. Sci., December 2007), or the earlier draft linked to at www.csse.monash.edu.au/~dld/David.Dowe.publications.html#DoweGardnerOppy2007 .

Possible areas of work include MML mixture modelling ( www.csse.monash.edu.au/~dld/Snob.html ), hierarchical mixture modelling, (hierarchical) mixture modelling with spatial correlation and image segmentation, time series segmentation, etc.

Other possible areas include regression, time series and autoregression models - and, of course, hybrid models more typically from the machine learning community, such as the abovementioned models and/or support vector machines (SVMs) in the leaves of discrete decision trees and/or graphs. All range of application areas (bioinformatics, etc.) are also welcomed.

As another more theoretical topic of possible interest, the Neyman-Scott (1948) problem turns the small sample problems of Maximum Likelihood on the Normal distribution into a statistical inconsistency. Issues of statistical (in)consistency both for Neyman-Scott-like problems (where the amount of data per parameter is bounded above) and under model misspecification are also possible areas of study. The available evidence to date suggests that statistical consistency and statistical invariance can only be retained in general by using a Bayesian approach.

For a more detailed (but not exhaustive) survey of possible areas of work, see, e.g., D. L. Dowe (2008a), ``Foreword re C. S. Wallace'', Computer J. (Oxford Univ. Press) [ http://dx.doi.org/10.1093/comjnl/bxm117 , http://comjnl.oxfordjournals.org/content/vol51/issue5 ], Christopher Stewart WALLACE (1933-2004) memorial special issue, pp523-560.

Successful applicants will be awarded an annual salary in the Research Fellow Level B range Aus$83,363 to Aus$98,994 p.a. (includes employer superannuation of 17%) plus a research support grant of up to Aus$50,000 per annum (p.a.), for five (5) years. The Fellowships are tenable only at Monash University and it is expected that most of the time will be spent at the relevant Monash department/school/centre - such as the Department of Computer Science in the Clayton School of Information Technology, Faculty of Information Technology with A./Prof. David Dowe (in Melbourne, Australia).

[Disclaimer/Qualifier: Eligible candidates can apply to work with a vast range of academics across a vast range of departments, schools and faculties at Monash, not just with yours truly.]

It is expected that up to four (4) awards will be made.

Successful applicants will be awarded an annual salary in the Research Fellow Level B range Aus$83,363 to Aus$98,994 p.a. (includes employer superannuation of 17%) plus a research support grant of up to Aus$50,000 per annum (p.a.), for five (5) years.

Duration: Five-year appointment.

Location: Monash University's Clayton campus, Melbourne, Australia.

Enquiries to:
A./Prof. David L. Dowe, david dot dowe at infotech dot monash.edu.au (www.csse.monash.edu.au/~dld/David.Dowe.publications.html) please by no later than some time in the first week of July 2009.

Location: Monash University's Clayton campus, Melbourne, Australia.