Gabriele Iommazzo

My academic interests primarily lie in the optimization of machine learning paradigms in mathematical programs. Recently, my focus has shifted towards convex optimization, particularly conditional gradient algorithms. This research explores their applications in diverse contexts, such as quantum nonlocality and optimization over intersecting sets.

📬 Contact

office
Room 3106 at ZIB
e-mail
homepage
giommazz.github.io/
languages
Italian, English, and French

🎓 Curriculum vitae

2022 to 2024
Researcher at ZIB
2021 to 2021
Research Fellow at University of Pisa
2017 to 2017
Research Intern at École polytechnique
Dec 2021
Ph.D. in Computer Science at and
Oct 2017
M.Sc. in Data Science and Business Informatics at University of Pisa

đź“ť Publications and preprints

Conference proceedings

  1. Iommazzo, G., D’Ambrosio, C., Frangioni, A., and Liberti, L. (2020). Learning to Configure Mathematical Programming Solvers by Mathematical Programming. Proceedings of the Learning and Intelligent Optimization Conference. DOI: 10.1007/978-3-030-53552-0_34 [arXiv]
    [BibTeX]
    @inproceedings{2020_IommazzoDAmbrosioFrangioniLiberti_Learningconfiguration,
  year = {2020},
  booktitle = {Proceedings of the Learning and Intelligent Optimization Conference},
  doi = {10.1007/978-3-030-53552-0_34},
  archiveprefix = {arXiv},
  eprint = {2401.05041},
  primaryclass = {math.OC},
  author = {Iommazzo, Gabriele and D'Ambrosio, Claudia and Frangioni, Antonio and Liberti, Leo},
  title = {Learning to Configure Mathematical Programming Solvers by Mathematical Programming}
}
  2. Liberti, L., Iommazzo, G., Lavor, C., and Maculan, N. (2020). A Cycle-based Formulation for Distance Geometry Problem. Proceedings of the 18th Cologne-Twente Workshop on Graphs and Combinatorial Optimization. DOI: 10.1007/978-3-030-63072-0_8
    [BibTeX]
    @inproceedings{2020_LibertiIommazzoLavorMaculan_Cycledistancegeometry,
  year = {2020},
  booktitle = {Proceedings of the 18th Cologne-Twente Workshop on Graphs and Combinatorial Optimization},
  doi = {10.1007/978-3-030-63072-0_8},
  author = {Liberti, Leo and Iommazzo, Gabriele and Lavor, Carlile and Maculan, Nelson},
  title = {A Cycle-based Formulation for Distance Geometry Problem}
}
  3. Iommazzo, G., D’Ambrosio, C., Frangioni, A., and Liberti, L. (2020). A Learning-based Mathematical Programming Formulation for the Automatic Configuration of Optimization Solvers. Proceedings of the International Conference on Machine Learning, Optimization and Data Science. DOI: 10.1007/978-3-030-64583-0_61 [arXiv]
    [BibTeX]
    @inproceedings{2021_IommazzoDAmbrosioFrangioniLiberti_Learningmathematicalprogramming,
  year = {2020},
  booktitle = {Proceedings of the International Conference on Machine Learning, Optimization and Data Science},
  doi = {10.1007/978-3-030-64583-0_61},
  archiveprefix = {arXiv},
  eprint = {2401.04237},
  primaryclass = {math.OC},
  author = {Iommazzo, Gabriele and D'Ambrosio, Claudia and Frangioni, Antonio and Liberti, Leo},
  title = {A Learning-based Mathematical Programming Formulation for the Automatic Configuration of Optimization Solvers}
}

Full articles

  1. Designolle, S., Iommazzo, G., Besançon, M., Knebel, S., Gelß, P., and Pokutta, S. (2023). Improved Local Models and New Bell Inequalities Via Frank-Wolfe Algorithms. Physical Review Research, 5(4). DOI: 10.1103/PhysRevResearch.5.043059 [arXiv] [slides] [code]
    [BibTeX]
    @article{2023_DesignolleEtAl_LocalmodelsBellinequalities,
  year = {2023},
  journal = {Physical Review Research},
  month = oct,
  volume = {5},
  number = {4},
  doi = {10.1103/PhysRevResearch.5.043059},
  archiveprefix = {arXiv},
  eprint = {2302.04721},
  primaryclass = {quant-ph},
  author = {Designolle, Sébastien and Iommazzo, Gabriele and Besançon, Mathieu and Knebel, Sebastian and Gelß, Patrick and Pokutta, Sebastian},
  title = {Improved Local Models and New Bell Inequalities Via Frank-Wolfe Algorithms},
  code = {https://github.com/ZIB-IOL/BellPolytopes.jl},
  slides = {https://www.pokutta.com/slides/20230808-tokyo-bell.pdf}
}
  2. Liberti, L., Iommazzo, G., Lavor, C., and Maculan, N. (2023). Cycle-based Formulations in Distance Geometry. Open Journal of Mathematical Optimization, 4(1). DOI: 10.5802/ojmo.18 [arXiv]
    [BibTeX]
    @article{2020_LibertiIommazzoLavorMaculan_Cycledistancegeometry_1,
  year = {2023},
  journal = {Open Journal of Mathematical Optimization},
  volume = {4},
  number = {1},
  doi = {10.5802/ojmo.18},
  archiveprefix = {arXiv},
  eprint = {2006.11523},
  primaryclass = {math.OC},
  author = {Liberti, Leo and Iommazzo, Gabriele and Lavor, Carlile and Maculan, Nelson},
  title = {Cycle-based Formulations in Distance Geometry}
}
  3. Iommazzo, G., D’Ambrosio, C., Frangioni, A., and Liberti, L. (2022). Algorithm Configuration Problem. In Encyclopedia of Optimization (pp. 1–8). DOI: 10.1007/978-3-030-54621-2_749-1 [arXiv]
    [BibTeX]
    @incollection{2022_IommazzoDAmbrosioFrangioniLiberti_Algorithmconfiguration,
  year = {2022},
  booktitle = {Encyclopedia of Optimization},
  pages = {1-8},
  doi = {10.1007/978-3-030-54621-2_749-1},
  archiveprefix = {arXiv},
  eprint = {2403.00898},
  primaryclass = {cs.AI},
  author = {Iommazzo, Gabriele and D'Ambrosio, Claudia and Frangioni, Antonio and Liberti, Leo},
  title = {Algorithm Configuration Problem}
}