Re: parallelization of potential operations

Marco Valtorta (mgv@cs.sc.edu)
Sat, 14 Nov 1998 16:58:16 -0500 (EST)

Dear Anders:
>
> Hi All,
>
> As has been mentioned by many researchers, a good way of exploiting
> parallelisme in multiprocessor computers when performing Bayesian
> inference is to have the table operations split up in independent
> subtasks.
>
> Does anyone know of any algorithms developed for performing
> multiplication, division, and marginalization of multiple dimensioned
> probability potentials or more general multiple dimensioned structures?

It seems to me that an obvious place to look would be papers on
parallel dynamic programming (in addition to the UAI paper by D'Ambrosio,
Fountain, and Li, which, despite the title that includes "early explorations"
is quite comprehensive, if memory does not fail me).

>
> Any replies will be appreciated.
>
> Thanks
> Anders L Madsen
>
> *--------------------------------------*
> | PhD-Student |
> | Anders L Madsen |
> *--------------------------------------*
> | Aalborg University |
> | Department of Computer Sience |
> | Frederik Bajers Vej 7C |
> | DK-9220 Aalborg |
> | Denmark |
> | Email : anders@cs.auc.dk |
> | Http : http://www.cs.auc.dk/~anders |
> *--------------------------------------*
>

Good luck!

Marco Valtorta, Associate Professor and Undergraduate Director
Department of Computer Science internet: mgv@usceast.cs.sc.edu
University of South Carolina tel.: (1)(803)777-4641 fax: -3767
Columbia, SC 29208, U.S.A. http://www.cs.sc.edu/~mgv/ tlx: 805038 USC