Change objective coefficient iteratively for SOCP with quadratic objective


#1

Hi all,

I am trying to write a program that will need to iteratively solve some problem with linear and second-order cone constraints and (minimize) quadratic objective function. To be more specific, at each iteration the coefficient of linear terms and constants in the objective will need to change, the constraints will stay the same.

From my understanding, since JuMP does not support quadratic objective for conic programming, I need to introduce a variable(t) to bound the objective form above and add (obj <= t) to the constraint. But it looks like I cannot change coefficients in the constraint or delete constraints from existing model.

I am wondering is there any way to address this issue?

I found line 24 and 25 on (https://groups.google.com/forum/#!topic/julia-opt/eUvWzFHaqxk) but looks like it didn’t change anything. Can anyone help?

Thanks!