Boro Sofranac

Mathematical Optimization
Quantagonia

I am interested in Mathematical Optimization and High-Performance Computing, with a focus on the intersection between the two fields. More specifically, I have been interested in developing (parallel) optimization algorithms that are capable of running efficiently on alternative hardware such as GPUs, FPGAs, ARM etc.

I hold a master's degree in Computational Science and Engineering from TU Munich and a bachelor's degree in Mechanical Engineering from the University of Montenegro.

I am affiliated with Quantagonia, where I work on developing the next-generation quantum algorithms.

Publications:

Preprints:

1. Bestuzheva et al. (2021). The SCIP Optimization Suite 8.0. Preprint. [arXiv]

Refereed conference proceedings:

1. Sofranac, B., Gleixner, A., & Pokutta, S. (2021). An Algorithm-Independent Measure of Progress for Linear Constraint Propagation. In L. D. Michel (Red), 27th International Conference on Principles and Practice of Constraint Programming (CP 2021) (bll 52:1-52:17). doi:10.4230/LIPIcs.CP.2021.52. [arXiv][code][slides][video]
2. Sofranac, B., Gleixner, A., & Pokutta, S. (2020). Accelerating Domain Propagation: An Efficient GPU-Parallel Algorithm over Sparse Matrices. 2020 IEEE/ACM 10th Workshop on Irregular Applications: Architectures and Algorithms (IA3), 1–11. doi:10.1109/IA351965.2020.00007. [arXiv][code][summary][slides][video]

Refereed journals:

1. Sofranac, B., Gleixner, A., & Pokutta, S. (2022). Accelerating domain propagation: An efficient GPU-parallel algorithm over sparse matrices. Parallel Computing, 109, 102874.  doi:​10.1016/j.parco.2021.102874 [arXiv][code][summary][slides][video]