Re: [microsound] Math Professors Solve 'Cocktail Party' Problem

[Balazs Gerofi <bgerofi@xxxxxxxxx>]

I think anything that a neural network does is represented by an algorithm.
They might mean that they proved some sort of an existence theorem.
Anyway, this "not reproducible neural network" sounds quite strange:)

Regards,
Balazs

On 8/27/06, David Powers <cyborgk@xxxxxxxxx> wrote:
>
> I believe what they mean is, the results cannot be reproduced, unless
> the neural network is properly trained. Without the training, which
> was possibly done on a trial-and-error basis, the neural network will
> not give successful results.
>
> The neural network itself is an algorithm, but there is no algorithm
> that you can run and get the successful results without the step of
> training the neural net.
>
> Although - I believe the training process itself should be able to be
> represented by an algorithm ...
>
> ~David
>
> On 8/27/06, Balazs Gerofi <bgerofi@xxxxxxxxx> wrote:
> > Hi,
> >
> > the article says:
> > "The computer we use is doing the work without an algorithmic program.
> It
> > uses a system called a neural net..."
> >
> > May I ask someone to explain why a neural network is not an algorithm?
> >
> >
> > Regards,
> > Balazs
> >
> > On 8/26/06, Graham Miller <grahammiller@xxxxxxxxxxxx> wrote:
> > >
> > > >>
> > >
> > > >> http://www.techweb.com/wire/security/192300009
> > >
> > >
> > >
> > > On 26-Aug-06, at 3:02 PM, isjtar junior wrote:
> > >
> > > > hi,
> > > >
> > > > that link doesn't seem to work, care to fix me a different one?
> > > > thanks
> > > >
> > > > isjtar
> > > > Op 25-aug-06 om 20:22 heeft Hans Erik Nilsson het volgende
> geschreven:
> > > >
> > > >> New adventures in field recordings: Paparazzi soundbites?
> > > >> [http://www.techweb.com/wire/security/192300009]
> > > >>
> > > >> /Cheers,
> > > >> /Hasse
> > > >>
> > > >> -- Mr. Hans Erik Nilsson | Voice: +46 73 6636331 | Standard
> > > >> disclaimer file active
> > > >> "There are many futures and only one status quo"  - Brian Eno
> > > >>
> > > >>
> ---------------------------------------------------------------------
> > > >> To unsubscribe, e-mail: microsound-unsubscribe@xxxxxxxxxxxxx
> > > >> For additional commands, e-mail: microsound-help@xxxxxxxxxxxxx
> > > >> website: http://www.microsound.org
> > > >>
> > > >>
> > > >
> ----------------------------------------------------------------------
> > > > -----------------------------
> > > > http://www.isjtar.org
> > > > http://www.societyofalgorithm.org
> > > >
> ----------------------------------------------------------------------
> > > > -----------------------------
> > > >
> > > >
> > > >
> ---------------------------------------------------------------------
> > > > To unsubscribe, e-mail: microsound-unsubscribe@xxxxxxxxxxxxx
> > > > For additional commands, e-mail: microsound-help@xxxxxxxxxxxxx
> > > > website: http://www.microsound.org
> > >
> > >
> > > ---------------------------------------------------------------------
> > > To unsubscribe, e-mail: microsound-unsubscribe@xxxxxxxxxxxxx
> > > For additional commands, e-mail: microsound-help@xxxxxxxxxxxxx
> > > website: http://www.microsound.org
> > >
> > >
> >
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: microsound-unsubscribe@xxxxxxxxxxxxx
> For additional commands, e-mail: microsound-help@xxxxxxxxxxxxx
> website: http://www.microsound.org
>
>