bject:
http://www.stanford.edu/~boyd/cvxbook/bv_cvxbook.pdf
Again thank you all for fruitful (at least for me :-P) discussion!
> Subject: Re: Consistent repeatable results for distributed ALS-WR recommender
> From: robin.e...@xense.co.uk
> Date: Tue, 25 Jun 2013 09:10:23 +0
y and repeatability? (for example we
> might want to track and compare the generated recommendation lists for
> different parameters, such as the number of features or number of iterations
> etc.)
> M.
>
>
>> Date: Mon, 24 Jun 2013 19:51:44 +0200
>> Subject: R
On Tue, Jun 25, 2013 at 12:44 AM, Michael Kazekin wrote:
> But doesn't alternation guarantee convexity?
No, the problem remains non-convex. At each step, where half the
parameters are fixed, yes that constrained problem is convex. But each
of these is not the same as the overall global problem be
Thanks for clarification, Owen!
> ALS starts from a random solution and this will result in a different
> solution. The overall problem is non-convex and the process will not
> necessarily converge to the same solution.
But doesn't alternation guarantee convexity?
> Randomness is a common feature o
Well, you know, the issue is there, whether we like it or not.
Maybe replication is enough, maybe not.
If there is a workshop on that issue, it's on the radar.
http://beamtenherrschaft.blogspot.com/2013/06/acm-recsys-2013-workshop-on.html
On Mon, Jun 24, 2013 at 6:36 PM, Sean Owen wrote:
> Yeah
Yeah this has gone well off-road.
ALS is not non-deterministic because of hardware errors or cosmic
rays. It's also nothing to do with floating-point round-off, or
certainly, that is not the primary source of non-determinism to
several orders of magnitude.
ALS starts from a random solution and th
algorithm
> implementation conserve (and why did authors added intentional
> non-deterministic component to implementation).
> > Date: Mon, 24 Jun 2013 14:43:59 -0700
> > Subject: Re: Consistent repeatable results for distributed ALS-WR
> recommender
> > From: dlie...@gmail.c
f the algorithm implementation
conserve (and why did authors added intentional non-deterministic component to
implementation).
> Date: Mon, 24 Jun 2013 14:43:59 -0700
> Subject: Re: Consistent repeatable results for distributed ALS-WR recommender
> From: dlie...@gmail.com
> To: user@mahout.a
There are definitely a
> > lot
> > > of such cases in Mahout.
> > >
> > > Another question is that afaik ALS-WR is deterministic by its
> inception,
> > so
> > > > I'm trying to understand the reasons (and I'm as
> >
> > > >
> > > > Almost all methods -- even deterministic ones -- will have a
> "credible
> > > > interval" of prediction simply because method assumptions do not hold
> > > 100%
> > > > in real life, rea
t; for
> > the specific implementation design.
> >
> > Thanks for a free lunch! ;)
> > Cheers,Mike.
> >
> > > Date: Mon, 24 Jun 2013 13:13:20 -0700
> > > Subject: Re: Consistent repeatable results for distributed ALS-WR
> > recommender
> >
so
> I'm trying to understand the reasons (and I'm assuming there are some) for
> the specific implementation design.
>
> Thanks for a free lunch! ;)
> Cheers,Mike.
>
> > Date: Mon, 24 Jun 2013 13:13:20 -0700
> > Subject: Re: Consistent repeatable results for dis
e effect on model credibility than
achieving ideal training cost.
> so I'm trying to understand the reasons (and I'm assuming there are some)
> for the specific implementation design.
>
> Thanks for a free lunch! ;)
> Cheers,Mike.
>
> > Date: Mon, 24 Jun 2013
ns (and I'm assuming there are some) for the
specific implementation design.
Thanks for a free lunch! ;)
Cheers,Mike.
> Date: Mon, 24 Jun 2013 13:13:20 -0700
> Subject: Re: Consistent repeatable results for distributed ALS-WR recommender
> From: dlie...@gmail.com
> To: user@mahout.
nistic ones in this context, and, therefore, more "useful". Also
see: "no free lunch theorem".
> > From: ted.dunn...@gmail.com
> > Date: Mon, 24 Jun 2013 20:46:43 +0100
> > Subject: Re: Consistent repeatable results for distributed ALS-WR
> recommender
> &
Thank you, Ted!
Any feedback on the usefulness of such functionality? Could it increase the
'playability' of the recommender?
> From: ted.dunn...@gmail.com
> Date: Mon, 24 Jun 2013 20:46:43 +0100
> Subject: Re: Consistent repeatable results for distributed ALS-WR rec
M.
> >
> >
> > > Date: Mon, 24 Jun 2013 19:51:44 +0200
> > > Subject: Re: Consistent repeatable results for distributed ALS-WR
> > recommender
> > > From: s...@apache.org
> > > To: user@mahout.apache.org
> > >
> > > The matrices of the
dation
> lists for different parameters, such as the number of features or number of
> iterations etc.)
> M.
>
>
> > Date: Mon, 24 Jun 2013 19:51:44 +0200
> > Subject: Re: Consistent repeatable results for distributed ALS-WR
> recommender
> > From: s...@apache.org
to track and compare the generated recommendation lists for
different parameters, such as the number of features or number of iterations
etc.)
M.
> Date: Mon, 24 Jun 2013 19:51:44 +0200
> Subject: Re: Consistent repeatable results for distributed ALS-WR recommender
> From: s...@apache
The matrices of the factorization are initalized randomly. If you fix the
random seed (would require modification of the code) you should get exactly
the same results.
Am 24.06.2013 13:49 schrieb "Michael Kazekin" :
> Hi!
> Should I assume that under same dataset and same parameters for factorizer
Hi!
Should I assume that under same dataset and same parameters for factorizer and
recommender I will get the same results for any specific user?
My current understanding that theoretically ALS-WR algorithm could guarantee
this, but I was wondering could be there any numeric method issues and/or
21 matches
Mail list logo