Unbalanced penalization for inequality constraints
Unbalanced penalization is an alternative way to represent inequality constraints without the need of using extra slack variables. We have shown that this method works well for the Traveling salesman problem and the bin packing problem using D-Wave Advantage and D-Wave hybrid solvers. Additionally, it outperforms the results obtained when using slack variables. It would be nice if you implement this method as a new type of constraint in Ocean.
[1] Montanez-Barrera, A., Maldonado-Romo, A., Willsch, D., & Michielsen, K. (2022). Unbalanced penalization: A new approach to encode inequality constraints of combinatorial problems for quantum optimization algorithms. 23–25. http://arxiv.org/abs/2211.13914
[2] Montanez-Barrera, J. A., Heuvel, P. van den, Willsch, D., & Michielsen, K. (2023). Improving Performance in Combinatorial Optimization Problems with Inequality Constraints: An Evaluation of the Unbalanced Penalization Method on D-Wave Advantage. http://arxiv.org/abs/2305.18757
Comments
Thank you for taking part in the community and for reaching out to us with your feature request.
I have passed this request along to the appropriate team for consideration.
We do not guarantee that the feature will be implemented, but letting us know which features would be useful helps us make more better decisions to help support our community.
An alternative approach is to contribute to the D-Wave Ocean Tools Library by implementing the feature and submitting it to the GitHub repository following the associated instructions.
Thank you, I created a new issue in dimod (https://github.com/dwavesystems/dimod/issues/1339).
Please sign in to leave a comment.