Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Lower (primal dual) start from upper solution candidate #35

Open
joaquimg opened this issue Mar 30, 2020 · 2 comments
Open

Lower (primal dual) start from upper solution candidate #35

joaquimg opened this issue Mar 30, 2020 · 2 comments

Comments

@joaquimg
Copy link
Owner

Obtain lower level primal and dual start values from upper level solution.

@NLaws
Copy link
Contributor

NLaws commented Jun 14, 2020

I have a large bilevel problem and I'm hoping that providing start values will aid in the solution process. Both my upper and lower levels are linear programs so it is easy to obtain some good start values. @joaquimg let me know if there is something I can do to help implement this feature.

@davide-f
Copy link
Contributor

@joaquimg I do have a large bilevel program as well and I believe that this improvement would be very important.
I believe that rather than an initial value for all dual variables, which is hard to find, it would be very interesting if it could be possibile to automatically initialize the complementarity constraints, so the corresponding binary variables or the SOS constraints.
When an entire set of primary variables is provided, it should be possible to calculate the slack variables of the constraints and hence set at least some of the binary conditions. I believe that this would significantly improve the performances of the models.

I have also added a comment on the JuMP package that could be of a support for this objective.
jump-dev/JuMP.jl#2299 (comment)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants