-
Notifications
You must be signed in to change notification settings - Fork 35
SB Parameters fine-tuning #105
Copy link
Copy link
Open
Labels
help wantedExtra attention is neededExtra attention is neededmodelThis PR adds a new optimization model to be solved with SBThis PR adds a new optimization model to be solved with SBquestionFurther information is requestedFurther information is requested
Description
Activity
Metadata
Metadata
Assignees
Labels
help wantedExtra attention is neededExtra attention is neededmodelThis PR adds a new optimization model to be solved with SBThis PR adds a new optimization model to be solved with SBquestionFurther information is requestedFurther information is requested
Currently, the parameters of the SB algorithm are hardcoded. Though they work quite well on unconstrained problems, we noticed a huge performance drop on constrained problems converted to QUBO/Ising formulations.
The purpose of this issue is to start a discussion and share ideas on how these parameters could be made scalable. First ideas include pre-determined sets of parameters for given problem typologies, machine learning for fine-tuning, grid-searches, ...
All help is welcome!