@inproceedings{MartinezRubioRouxCriscitielloetal.2023, author = {Mart{\´i}nez-Rubio, David and Roux, Christophe and Criscitiello, Christopher and Pokutta, Sebastian}, title = {Accelerated Riemannian Min-Max Optimization Ensuring Bounded Geometric Penalties}, booktitle = {Proceedings of Optimization for Machine Learning (NeurIPS Workshop OPT 2023)}, year = {2023}, language = {en} } @article{DesignolleVertesiPokutta2024, author = {Designolle, S{\´e}bastien and V{\´e}rtesi, Tam{\´a}s and Pokutta, Sebastian}, title = {Symmetric multipartite Bell inequalities via Frank-Wolfe algorithms}, volume = {109}, journal = {Physics Review A}, number = {2}, doi = {10.1103/PhysRevA.109.022205}, year = {2024}, abstract = {In multipartite Bell scenarios, we study the nonlocality robustness of the Greenberger-Horne-Zeilinger (GHZ) state. When each party performs planar measurements forming a regular polygon, we exploit the symmetry of the resulting correlation tensor to drastically accelerate the computation of (i) a Bell inequality via Frank-Wolfe algorithms and (ii) the corresponding local bound. The Bell inequalities obtained are facets of the symmetrized local polytope and they give the best-known upper bounds on the nonlocality robustness of the GHZ state for three to ten parties. Moreover, for four measurements per party, we generalize our facets and hence show, for any number of parties, an improvement on Mermin's inequality in terms of noise robustness. We also compute the detection efficiency of our inequalities and show that some give rise to the activation of nonlocality in star networks, a property that was only shown with an infinite number of measurements.}, language = {en} } @inproceedings{ChmielaKhalilGleixneretal.2021, author = {Chmiela, Antonia and Khalil, Elias B. and Gleixner, Ambros and Lodi, Andrea and Pokutta, Sebastian}, title = {Learning to Schedule Heuristics in Branch and Bound}, booktitle = {Thirty-fifth Conference on Neural Information Processing Systems, NeurIPS 2021}, arxiv = {http://arxiv.org/abs/2103.10294}, year = {2021}, abstract = {Primal heuristics play a crucial role in exact solvers for Mixed Integer Programming (MIP). While solvers are guaranteed to find optimal solutions given sufficient time, real-world applications typically require finding good solutions early on in the search to enable fast decision-making. While much of MIP research focuses on designing effective heuristics, the question of how to manage multiple MIP heuristics in a solver has not received equal attention. Generally, solvers follow hard-coded rules derived from empirical testing on broad sets of instances. Since the performance of heuristics is instance-dependent, using these general rules for a particular problem might not yield the best performance. In this work, we propose the first data-driven framework for scheduling heuristics in an exact MIP solver. By learning from data describing the performance of primal heuristics, we obtain a problem-specific schedule of heuristics that collectively find many solutions at minimal cost. We provide a formal description of the problem and propose an efficient algorithm for computing such a schedule. Compared to the default settings of a state-of-the-art academic MIP solver, we are able to reduce the average primal integral by up to 49\% on a class of challenging instances.}, language = {en} } @inproceedings{ChmielaGleixnerLichockietal.2023, author = {Chmiela, Antonia and Gleixner, Ambros and Lichocki, Pawel and Pokutta, Sebastian}, title = {Online Learning for Scheduling MIP Heuristics}, volume = {13884}, booktitle = {Integration of Constraint Programming, Artificial Intelligence, and Operations Research}, publisher = {Springer, Cham}, doi = {10.1007/978-3-031-33271-5_8}, pages = {114 -- 123}, year = {2023}, abstract = {Mixed Integer Programming (MIP) is NP-hard, and yet modern solvers often solve large real-world problems within minutes. This success can partially be attributed to heuristics. Since their behavior is highly instance-dependent, relying on hard-coded rules derived from empirical testing on a large heterogeneous corpora of benchmark instances might lead to sub-optimal performance. In this work, we propose an online learning approach that adapts the application of heuristics towards the single instance at hand. We replace the commonly used static heuristic handling with an adaptive framework exploiting past observations about the heuristic's behavior to make future decisions. In particular, we model the problem of controlling Large Neighborhood Search and Diving - two broad and complex classes of heuristics - as a multi-armed bandit problem. Going beyond existing work in the literature, we control two different classes of heuristics simultaneously by a single learning agent. We verify our approach numerically and show consistent node reductions over the MIPLIB 2017 Benchmark set. For harder instances that take at least 1000 seconds to solve, we observe a speedup of 4\%.}, language = {en} } @inproceedings{GasseBowlyCappartetal.2022, author = {Gasse, Maxime and Bowly, Simon and Cappart, Quentin and Charfreitag, Jonas and Charlin, Laurent and Ch{\´e}telat, Didier and Chmiela, Antonia and Dumouchelle, Justin and Gleixner, Ambros and Kazachkov, Aleksandr M. and Khalil, Elias and Lichocki, Pawel and Lodi, Andrea and Lubin, Miles and Maddison, Chris J. and Christopher, Morris and Papageorgiou, Dimitri J. and Parjadis, Augustin and Pokutta, Sebastian and Prouvost, Antoine and Scavuzzo, Lara and Zarpellon, Giulia and Yang, Linxin and Lai, Sha and Wang, Akang and Luo, Xiaodong and Zhou, Xiang and Huang, Haohan and Shao, Shengcheng and Zhu, Yuanming and Zhang, Dong and Quan, Tao and Cao, Zixuan and Xu, Yang and Huang, Zhewei and Zhou, Shuchang and Binbin, Chen and Minggui, He and Hao, Hao and Zhiyu, Zhang and Zhiwu, An and Kun, Mao}, title = {The Machine Learning for Combinatorial Optimization Competition (ML4CO): results and insights}, booktitle = {Proceedings of Conference on Neural Information Processing Systems}, year = {2022}, language = {en} } @article{CardereraPokuttaSchuetteetal.2025, author = {Carderera, Alejandro and Pokutta, Sebastian and Sch{\"u}tte, Christof and Weiser, Martin}, title = {An efficient first-order conditional gradient algorithm in data-driven sparse identification of nonlinear dynamics to solve sparse recovery problems under noise}, volume = {470}, journal = {Journal of Computational and Applied Mathematics}, doi = {10.1016/j.cam.2025.116675}, year = {2025}, abstract = {Governing equations are essential to the study of nonlinear dynamics, often enabling the prediction of previously unseen behaviors as well as the inclusion into control strategies. The discovery of governing equations from data thus has the potential to transform data-rich fields where well-established dynamical models remain unknown. This work contributes to the recent trend in data-driven sparse identification of nonlinear dynamics of finding the best sparse fit to observational data in a large library of potential nonlinear models. We propose an efficient first-order Conditional Gradient algorithm for solving the underlying optimization problem. In comparison to the most prominent alternative framework, the new framework shows significantly improved performance on several essential issues like sparsity-induction, structure-preservation, noise robustness, and sample efficiency. We demonstrate these advantages on several dynamics from the field of synchronization, particle dynamics, and enzyme chemistry.}, language = {en} } @inproceedings{LasbyZimmerPokuttaetal.2025, author = {Lasby, Mike and Zimmer, Max and Pokutta, Sebastian and Schultheis, Erik}, title = {Compressed sparse tiles for memory-efficient unstructured and semi-structured sparsity}, booktitle = {Proceedings of the ICLR Workshop on Sparsity in LLMs}, year = {2025}, language = {en} } @inproceedings{HaaseHanelPokutta2025, author = {Haase, Jennifer and Hanel, Paul H. P. and Pokutta, Sebastian}, title = {S-DAT: a multilingual, GenAI-driven framework for automated divergent thinking assessment}, volume = {8}, booktitle = {Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society (AIES)}, pages = {1194 -- 1205}, year = {2025}, language = {en} } @article{WirthPenaPokutta2025, author = {Wirth, Elias and Pe{\~n}a, Javier and Pokutta, Sebastian}, title = {Fast convergence of Frank-Wolfe algorithms on polytopes}, journal = {INFORMS Journal on Mathematics of Operations Research}, doi = {10.1287/moor.2024.0580}, year = {2025}, language = {en} } @article{WoodstockPokutta2025, author = {Woodstock, Zev and Pokutta, Sebastian}, title = {Splitting the conditional gradient algorithm}, volume = {35}, journal = {SIAM Journal on Optimization}, number = {1}, doi = {10.1137/24M1638008}, pages = {347 -- 368}, year = {2025}, language = {en} } @article{GonnermannMuellerHaaseLeinsetal.2026, author = {Gonnermann-M{\"u}ller, Jana and Haase, Jennifer and Leins, Nicolas and Igel, Moritz and Fackeldey, Konstantin and Pokutta, Sebastian}, title = {FACET: Multi-Agent AI Supporting Teachers in Scaling Differentiated Learning for Diverse Students}, journal = {arXiv}, arxiv = {http://arxiv.org/abs/2601.22788}, doi = {https://arxiv.org/abs/2601.22788}, year = {2026}, abstract = {Classrooms are becoming increasingly heterogeneous, comprising learners with diverse performance and motivation levels, language proficiencies, and learning differences such as dyslexia and ADHD. While teachers recognize the need for differentiated instruction, growing workloads create substantial barriers, making differentiated instruction an ideal that is often unrealized in practice. Current AI educational tools, which promise differentiated materials, are predominantly student-facing and performance-centric, ignoring other aspects that shape learning outcomes. We introduce FACET, a teacher-facing multi-agent framework designed to address these gaps by supporting differentiation that accounts for motivation, performance, and learning differences. Developed with educational stakeholders from the outset, the framework coordinates four specialized agents, including learner simulation, diagnostic assessment, material generation, and evaluation within a teacher-in-the-loop design. School principals (N = 30) shaped system requirements through participatory workshops, while in-service K-12 teachers (N = 70) evaluated material quality. Mixed-methods evaluation demonstrates strong perceived value for inclusive differentiation. Practitioners emphasized both the urgent need arising from classroom heterogeneity and the importance of maintaining pedagogical autonomy as a prerequisite for adoption. We discuss implications for future school deployment and outline partnerships for longitudinal classroom implementation.}, language = {en} } @article{GonnermannMuellerHaaseFackeldeyetal.2025, author = {Gonnermann-M{\"u}ller, Jana and Haase, Jennifer and Fackeldey, Konstantin and Pokutta, Sebastian}, title = {FACET: Teacher-Centred LLM-Based Multi-Agent Systems-Towards Personalized Educational Worksheets}, arxiv = {http://arxiv.org/abs/2508.11401}, year = {2025}, abstract = {The increasing heterogeneity of student populations poses significant challenges for teachers, particularly in mathematics education, where cognitive, motivational, and emotional differences strongly influence learning outcomes. While AI-driven personalization tools have emerged, most remain performance-focused, offering limited support for teachers and neglecting broader pedagogical needs. This paper presents the FACET framework, a teacher-facing, large language model (LLM)-based multi-agent system designed to generate individualized classroom materials that integrate both cognitive and motivational dimensions of learner profiles. The framework comprises three specialized agents: (1) learner agents that simulate diverse profiles incorporating topic proficiency and intrinsic motivation, (2) a teacher agent that adapts instructional content according to didactical principles, and (3) an evaluator agent that provides automated quality assurance. We tested the system using authentic grade 8 mathematics curriculum content and evaluated its feasibility through a) automated agent-based assessment of output quality and b) exploratory feedback from K-12 in-service teachers. Results from ten internal evaluations highlighted high stability and alignment between generated materials and learner profiles, and teacher feedback particularly highlighted structure and suitability of tasks. The findings demonstrate the potential of multi-agent LLM architectures to provide scalable, context-aware personalization in heterogeneous classroom settings, and outline directions for extending the framework to richer learner profiles and real-world classroom trials.}, language = {en} }