3,553 reputation
12051
bio website regularize.wordpress.com
location Braunschweig, Germany
age 36
visits member for 4 years, 10 months
seen Jul 29 at 8:47

Professor at TU Braunschweig.

Area: Inverse problems, regularization theory, applied functional analysis, mathematical image processing.


Jul
24
comment How to calculate $det(X^TX)$ efficiently, update one column of X each time
Not a good fit for this side - I suggest scicomp.stackexchange.com. Anyway: Why not do a factorization of $X_1$ (LU or QR) and down- and update that?
Jul
24
comment 9-point stencil “equivalent” for advection equation
This question should go well on scicomp.stackexchange.com.
Jul
17
comment question about $TGV^2$ space
Added something on the interpretation with Radon measures…
Jul
17
comment Is this has anything to do with Riesz representation?
Nope, not linear since flipping the sign of my won't flip the sign of the value.
Jul
17
comment Is this has anything to do with Riesz representation?
One more thing: Your functional is not linear but only positively homogeneous, so what kind of result do you expect?
Jul
17
comment Is this has anything to do with Riesz representation?
What form of Sobolev norm are you using, i. e. how are the sup-norm of the function and it's derivative combined?
May
20
comment Projecting a convex partition onto a convex set
Err, sorry - of course, nonempty $X$ it should be…
May
20
comment Projecting a convex partition onto a convex set
$X$ could be empty or have $n-1$ points... You need some more conditions.
May
19
comment Splines linearly independent
What do you mean by "its unique smooth extension"?
May
14
comment Which way for reading the proofs?
@fanzheng I heard a similar quote about a famous optimizer "I don't read papers, I write them."
Apr
23
comment l1 Quadratic Programming
Noting that the above problem has all these guys as variables: Isn't that in standard for already?
Apr
23
comment A question involving Mazur's Lemma
Well, the degenerate case $y_n = x_n$ does not work so some assumption on the convex combination is needed.
Apr
14
comment 2, 3, and 4 (a possible fixed point result ?)
As far as I understood, the BGK (Browder-Göhde/Göbel-Kirk) fixed point theorem states that every non-expansive self-mapping on a non-empty, closed and convex subset of a uniformly convex Banach space has a fixed point.
Mar
27
comment Hadamard / matrix product adjoint
Ok, I see now. I think a better fit for this question would be scicomp.stackexchange.com. Anyway: The adjoint is defined by $\langle A^* y,x\rangle = \langle y,Ax\rangle$. You do adjoints one by one from outside to inside.
Mar
26
comment Hadamard / matrix product adjoint
Sorry, but this does not make sense. What is the Hadamard product of the matrix S and the vector Fx?
Mar
26
comment Hadamard / matrix product adjoint
Err... So x is a matrix?
Mar
26
comment Hadamard / matrix product adjoint
I could not parse your definition of the operator A. If D and S are matrices, do you build the Hadamard product with the matrix of the Fourier transform?
Mar
26
comment Convergence in energy of bounded (semi)subharmonic functions
Sorry for the late reply: I guess the best thing would be to write an answer yourself so that this is kept for the records.
Mar
17
comment Fredholm integral with functions constrained to [0;1]
For some reason I confused $g$ and $f$. Of course $f$ should be the unknown… Corrected.
Mar
16
comment Fredholm integral with functions constrained to [0;1]
The most simple thing that comes to mind is the projected gradient method, see my update.