Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: the "set paradigm" is harmful



Arnold,

please see below my interpolated comments
to your remarks.

On 12 Feb 2009 at 13:17, Arnold Neumaier wrote:

> Svetoslav Markov schrieb:
> > 
> >  Surely, MR-presented intervals, resp. approximate numbers, 
> > admit probabilistic interpretation, e.g. midpoint can 
> > be mean value, in Gaussian distributions radius can be 
> > standard deviation or half-width of confidence interval, etc. 
> > (Note that the guaranteed containment property can be still there,
> > say within the 95% of probability). 
> 
> This is not true.
> 
> if |x-x_0|<=sigma with probability 95%  and
> |y-y_0|<=sigma' with probability 95%, there is no guarantee that
> the product x*y will lie in the product of the corresponding
> intervals with 95%.
> 

I wrote "can be", not "is". What I say is true for widely used
probability operations such as addition and multiplication
by scalars. 

> 
> > Note that
> > MR-form intervals incorporate the set paradigm --- up to the 
> > special cases (oo, a], [b, oo). 
> 
> Not if, as you suggest, the radius is represented by 2 digits only.
> 

1-2-3 digit representation is suggested for NARROW intervals.
Midrad representation/computation is practically useful for narrow
intervals, the intervals that model approximate numbers.

> For simplicity, assume a decimal representation, and consider the
> interval xx=[0,202]. Its conversion to midrad form is ambiguous and 
> overestimates. The three centered intervals <mid=9.2e1,rad=1.1e2>,
> <mid=1.01e2,rad=1.1e2>, and <mid=1.1e2,rad=1.1e2> are among the many
> minimal centered intervals with 2 digit radius containing xx,
> corresponding to the intervals [-18,202], [-9,211] and [0,220].
> 
> One loses almost 9% of the accuracy (1.1/1.01=1.0891...), and all
> apart from the last version don't even preserve the sign information 
> inherent in the infsup presentation.
> 
> Also, the cost of a rigorous midrad interval arithmetic (including 
> rounding error control) is twice that of infsup arithmetic.
> Proponents of midrad arithmetic should first remove this disatvantage.
> 

Here it seems that you forget about the 2-digit mantissa
representation!

But I am glad that you admit that in real arithmetic the cost is same.

> 
> > That is why MR-form is even more
> > important than sup-inf form. 
> 
> That is why the MR-form is useless for a standard.
> Anyone interested in using the midrad form can easily convert it
> at the beginning and end of a computation with the functions proposed
> in the Vienna Proposal. More is not needed.
> 

IMO it is needed for fast computations with narrow intervals.

> 
> > MR-form is more convenient than the
> > sup-inf form in many applications (possibly for Taylor polynomials
> > as well).
> 
> If this is so, who is currently using the midrad form for executing 
> interval operations in some of these many applications?
> 
> Intlab provides an intermediate midrad representation
> in a fast matrix multiply option, but at the expense of an 
> overestimation factor of 1.5 for the results widths.
> To get rid of this factor would make the midrad representation
> slower than the standard version!
> 

again: narrow intervals & 2-digit mantissa!

> 
>   Some authors often use this form as an intrinsic form in
> > computations/proofs and then translate results in sup-inf form
> > (notably J. Rohn). 
> 
> Nobody says that mid and rad are not useful theoretical tools.
> But the standard is about standardizing implementations.
> 
> Where is the fast and much-used rigorous implementation of
> midrad arithmetic that people use in their applications?
> I haven't heard of any.
> 

who knows, may  there will soon appear patents about midrad
computations ...

> 
> > Conclusion is: MR-presentation of intervals deserves (at least) equal
> > consideration in the standard as the sup-inf form.
> 
> This conclusion is completely unsupported by the facts.

facts? I cannot see them.
 

Regards,

Svetoslav