[SciPy-User] optimization routines can not handle infinity values
Sebastian Walter
sebastian.walter@gmail....
Wed Sep 15 10:28:56 CDT 2010
you said you formulated your problem as follows:
f1(x) = f(x) for M(x) pos semi def
f1(x) = Inf otherwise
I don't see any indication that anything goes to infinity when the sequence
of iterates approaches a boundary.
It seems to me that what you are describing is the behavior of a barrier
function which is used in interior point methods, i.e. of the form
min_x f(x) - a log( g(M(x))
where g(M(x)) is a sufficiently smooth function in x and g(M(x)) = 0 if M(x)
is not positive definite
and a>0. Is this what you are doing?
Sebastian
On Wed, Sep 15, 2010 at 12:52 PM, Enrico Avventi <eavventi@yahoo.it> wrote:
> my function tends to infinity as you approach the boundary, as it is the
> case for all convex functions that can be defined only inside an open convex
> set.
>
>
> On Wed, Sep 15, 2010 at 12:18 PM, Sebastian Walter <
> sebastian.walter@gmail.com> wrote:
>
>> I can't quite follow the reasoning that another line search would solve
>> the problem.
>> If the current iterate is at the boundary and the new search direction
>> points into the infeasible set, then a line search is unlikely to help.
>>
>> Example:
>> min_x dot([2,1],x)
>> s.t. x >= 0
>>
>> the minimizer is x_* = [0,0]. If the current iterate x_k is at [3,0], then
>> the new search direction woud be
>> s_k = -[2,1] which points directly into the infeasible set which would
>> yield infty.
>>
>> I'm not sure what the best approach for your problem is.
>> Since SDP is still an active topic of research one could conjecture that
>> possibly it's not as easy as you think.
>> If you find a good solution I'd be happy to hear about it.
>>
>> regards,
>> Sebastian
>>
>>
>>
>>
>> On Wed, Sep 15, 2010 at 11:19 AM, Matthieu Brucher <
>> matthieu.brucher@gmail.com> wrote:
>>
>>> Hi,
>>>
>>> The line search used in scipy is based on Wolfe-Powell rules and the
>>> search for appropriate values is done by interpolation. This is why it
>>> cannot be used for your kind of problems. All Wolfe-Powel line searches I've
>>> found in the litterature are based on such interpolations.
>>> It could be changed to fit your purpose, but it would be slower for
>>> 99.99% of the other optimizations. So you may add some additional
>>> constraints here.
>>>
>>> Matthieu
>>>
>>> 2010/9/15 Enrico Avventi <eavventi@yahoo.it>
>>>
>>> Hi Matthieu,
>>>>
>>>> thanx for the reply. as far as i know it shouldn't be a problem at all
>>>> in theory. in fact convergence theorems for newton methods and, in general,
>>>> descent method do not require the function to be defined everywhere. you
>>>> need just a function defined in an open convex set that has compact sublevel
>>>> sets - e.g f(x) = x - log(x). this is exactly my situation.
>>>> it is just a matter of slightly changing the line search method to
>>>> reject steps that would lead to an infinite value.
>>>> i will look at how it is implemented in scipy and see if it can be fixed
>>>> easily.
>>>>
>>>> /Enrico
>>>>
>>>>
>>>> On Tue, Sep 14, 2010 at 5:51 PM, Matthieu Brucher <
>>>> matthieu.brucher@gmail.com> wrote:
>>>>
>>>>> Hi,
>>>>>
>>>>> There are two categories of contraints optimizations:
>>>>> - you can evaluate the function outside the constraints
>>>>> - you cannot evaluate the function outsde the constraints.
>>>>>
>>>>> If the first one can be handled by more general algorithms providing
>>>>> some tricks, you cannot use them for the second one. Your problem is clearly
>>>>> a second category problem, so you must use appropriate algorithms (which may
>>>>> not be available in scipy directly, you may want to check OpenOpt).
>>>>>
>>>>> It's not a problem of routines, it's a problem of appropriate
>>>>> algorithms.
>>>>>
>>>>> Matthieu
>>>>>
>>>>> 2010/9/14 enrico avventi <eavventi@yahoo.it>
>>>>>
>>>>>> hello all,
>>>>>>
>>>>>> i am trying out some of the optimization routines for a problem of
>>>>>> mine that is on the form:
>>>>>>
>>>>>> min f(x)
>>>>>>
>>>>>> s.t M(x) is positive semidefinite
>>>>>>
>>>>>> where f is strictly convex in the feasible region with compact
>>>>>> sublevel sets, M is linear and takes value in some subspace of hermitian
>>>>>> matrices.
>>>>>>
>>>>>> the problem is convex but the costraint can not be handled directly by
>>>>>> any of the optimization routines in scipy. So i choose to change it to an
>>>>>> uncostrained problem with objective function:
>>>>>>
>>>>>> f1(x) = f(x) for M(x) pos semi def
>>>>>> f1(x) = Inf otherwise
>>>>>>
>>>>>> the problem is that it seems the routines can not handle the infinity
>>>>>> values correctly.
>>>>>>
>>>>>> Some of the routines (fmin_cg comes to mind) wants to check the
>>>>>> gradient at points where the objective function is infinite. Clearly in such
>>>>>> cases the gradient is not defined - i.e the calculations fail - and the
>>>>>> algorithm terminates.
>>>>>>
>>>>>> Others (like fmin_bfgs) strangely converge to a point where the
>>>>>> objective is infinite despite the fact that the initial point was not.
>>>>>>
>>>>>> Do you have any suggestion to fix this problem?
>>>>>>
>>>>>> regards,
>>>>>>
>>>>>> Enrico
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> SciPy-User mailing list
>>>>>> SciPy-User@scipy.org
>>>>>> http://mail.scipy.org/mailman/listinfo/scipy-user
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Information System Engineer, Ph.D.
>>>>> Blog: http://matt.eifelle.com
>>>>> LinkedIn: http://www.linkedin.com/in/matthieubrucher
>>>>>
>>>>> _______________________________________________
>>>>> SciPy-User mailing list
>>>>> SciPy-User@scipy.org
>>>>> http://mail.scipy.org/mailman/listinfo/scipy-user
>>>>>
>>>>>
>>>>
>>>> _______________________________________________
>>>> SciPy-User mailing list
>>>> SciPy-User@scipy.org
>>>> http://mail.scipy.org/mailman/listinfo/scipy-user
>>>>
>>>>
>>>
>>>
>>> --
>>> Information System Engineer, Ph.D.
>>> Blog: http://matt.eifelle.com
>>> LinkedIn: http://www.linkedin.com/in/matthieubrucher
>>>
>>> _______________________________________________
>>> SciPy-User mailing list
>>> SciPy-User@scipy.org
>>> http://mail.scipy.org/mailman/listinfo/scipy-user
>>>
>>>
>>
>> _______________________________________________
>> SciPy-User mailing list
>> SciPy-User@scipy.org
>> http://mail.scipy.org/mailman/listinfo/scipy-user
>>
>>
>
> _______________________________________________
> SciPy-User mailing list
> SciPy-User@scipy.org
> http://mail.scipy.org/mailman/listinfo/scipy-user
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.scipy.org/pipermail/scipy-user/attachments/20100915/3ae85b7d/attachment-0001.html
More information about the SciPy-User
mailing list