What we are interested in
We study algorithms and theory for large-scale continuous optimization, with a focus on first-order methods for problems arising in modern machine learning. Our goal is to design methods that are provably convergent, efficient, and robust in practice. We work on projection-free conditional-gradient (Frank–Wolfe) variants for large-scale constrained problems, online and stochastic optimization, and robust and min–max formulations that capture distribution shift and adversarial settings. Furthermore, we study optimization for deep learning, including developing principled optimizers for training deep neural networks, methods for federated learning, and techniques that make training and inference more efficient and sustainable.
Researchers
Doctoral candidates
Undergraduate and Phase I Students
Former members
- Seta Rakotomandimby (2025)
- David Martínez-Rubio (2022–2024)
- Zev Woodstock (2022–2024)
- Elias Wirth (2020–2024)
- Mathieu Besançon (2021–2023)
- Paco Criado (2021–2022)
- Alejandro Carderera (2018–2021)
- Cyrille Combettes (2018–2021)
- Thomas Kerdreux (2020–2021)
- Cristóbal Guzmán (2012–2015)
Visitors
- Manuella Nakam Yopdup (September 2025–)
- Mathieu Besançon (February 2025)
External Collaborators
Talks and posters
Conference and workshop talks
- Jul 2026
- The Agentic Researcher: A Practical Guide to AI-Assisted Research in Mathematics and Machine Learning by Nico Pelleriti
AI4Research Workshop, Seoul - Jun 2026
- Exact Solutions for the Graph Isomorphism with Convex Mixed Integer Optimization by Wenjie Xiao
SIAM conference on optimization (SIAMOP), Edinburgh - May 2026
- The Graph Isomorphism Problem From a Mixed-Integer Convex Optimization Perspective by Wenjie Xiao
23rd CPAIOR Conference, Rabat - Jul 2025
- Online Implicit Riemannian Optimization by Christophe Roux
8th International Conference on Continuous Optimization (ICCOPT), Los Angeles - Jul 2025
- Corrective Frank-Wolfe: Unifying and Extending Correction Steps by Jannis Halbey
22nd Conference on Advances in Continuous Optimization (EUROPT), Southampton - May 2025
- Implicit Riemannian Optimism with Applications to Min-Max Problems by Christophe Roux
Foundations and Frontiers: Interdisciplinary Perspectives on Mathematical Optimization, Nihonbashi
Research seminar talks
- Sep 2026
- Curvature-Dependent Lower Bounds for Frank-Wolfe by Jannis Halbey
GHOST Research Seminar (GHOST), Grenoble - Mar 2026
- Lower Bounds for Frank-Wolfe on Strongly Convex Sets by Christophe Roux
SIERRA Seminar at INRIA Paris, Paris - Mar 2025
- Decomposition-Invariant Frank-Wolfe Method in Mixed-Integer Convex Optimization with an Application to Graph Isomorphism by Wenjie Xiao
IOL Research Seminar (IOL), Berlin - Apr 2024
- Bounding Geometric Penalties in Riemannian Optimization by Christophe Roux
CISPA Research seminar, Saarbrücken - Nov 2023
- Bounding Geometric Penalties in First-Order Riemannian Optimization by Christophe Roux
Modern Methods in Applied Stochastics and Nonparametric Statistics, Berlin
Poster presentations
- Jul 2026
- Recent Improvements on the Integer Frank-Wolfe Framework by Deborah Hendrych
ICOML Conference, Taipei - Jul 2026
- Curvature-Dependent Lower Bounds for Frank-Wolfe by Jannis Halbey
8th DOxML Conference, Ookayama - Jul 2026
- Recent Improvements on the Integer Frank-Wolfe Framework by Deborah Hendrych
8th DOxML Conference, Ookayama - Jul 2026
- The Agentic Researcher: A Practical Guide to AI-Assisted Research in Mathematics and Machine Learning by Nico Pelleriti
AI4Math Workshop, Seoul - Jul 2026
- The Agentic Researcher: A Practical Guide to AI-Assisted Research in Mathematics and Machine Learning by Nico Pelleriti
AI4Research Workshop, Seoul - Jul 2026
- Lower Bounds for Frank-Wolfe on Strongly Convex Sets by Jannis Halbey
43rd International Conference on Machine Learning (ICML), Seoul - Dec 2025
- Efficient Quadratic Corrections for Frank-Wolfe Algorithms by Jannis Halbey
39th conference on neural information processing systems (NeurIPS), San Diego - Jul 2025
- Implicit Riemannian Optimism with Applications to Min-Max Problems by Christophe Roux
42nd International Conference on Machine Learning (ICML), Vancouver - Jul 2025
- Secant Line Search for Frank-Wolfe Algorithms by Deborah Hendrych
42nd International Conference on Machine Learning (ICML), Vancouver - Jul 2025
- Approximating Latent Manifolds in Neural Networks via Vanishing Ideals by Nico Pelleriti and Max Zimmer
42nd International Conference on Machine Learning (ICML), Vancouver - May 2025
- Accelerated Riemannian Min-Max Optimization Ensuring Bounded Geometric Penalties by Christophe Roux
28th AISTATS Conference, Phuket - Jul 2024
- Convergence and Trade-Offs in Riemannian Gradient Descent and Riemannian Proximal Point by Christophe Roux
41st International Conference on Machine Learning (ICML), Vienna - Mar 2023
- Accelerated Riemannian Min-Max Optimization Ensuring Bounded Geometric Penalties by Christophe Roux
Workshop on Optimization and Machine Learning, Waischenfeld
Publications and preprints
Preprints and reports
- Sharma, U., Goel, K., Dua, A., Pokutta, S., and Woodstock, Z. (2025). A note on asynchronous projective splitting in Julia.
[URL]
[BibTeX]
@misc{2025_SharmaGoelDuaPokuttaWoodstock_AsyncProx, year = {2025}, month = apr, url = {https://zevwoodstock.github.io/media/publications/asyncprox.pdf}, author = {Sharma, Utkarsh and Goel, Kashish and Dua, Aryan and Pokutta, Sebastian and Woodstock, Zev}, title = {A note on asynchronous projective splitting in Julia} } - Braun, G., and Pokutta, S. (2016). An efficient high-probability algorithm for Linear Bandits.
[arXiv]
[BibTeX]
@misc{2016_BraunPokutta_Highprobabilitylinearbandits, archiveprefix = {arXiv}, eprint = {1610.02072}, arxiv = {arXiv:1610.02072}, primaryclass = {cs.DS}, author = {Braun, Gábor and Pokutta, Sebastian}, title = {An efficient high-probability algorithm for Linear Bandits}, date = {2016-10-06} } - Combettes, C., Spiegel, C., and Pokutta, S. (2020). Projection-Free Adaptive Gradients for Large-Scale Optimization.
[arXiv]
[summary]
[code]
[BibTeX]
@misc{2020_CombettesSpiegelPokutta_Projectionfreeadaptivegradients, archiveprefix = {arXiv}, eprint = {2009.14114}, arxiv = {arXiv:2009.14114}, primaryclass = {math.OC}, author = {Combettes, Cyrille and Spiegel, Christoph and Pokutta, Sebastian}, title = {Projection-Free Adaptive Gradients for Large-Scale Optimization}, code = {https://github.com/ZIB-IOL/StochasticFrankWolfe}, summary = {https://pokutta.com/blog/research/2020/10/21/adasfw.html}, date = {2020-09-29} } - Braun, G., and Pokutta, S. (2021). Dual Prices for Frank–Wolfe Algorithms.
[arXiv]
[BibTeX]
@misc{2021_BraunPokutta_DualpricesFrankwolfe, archiveprefix = {arXiv}, eprint = {2101.02087}, arxiv = {arXiv:2101.02087}, primaryclass = {math.OC}, author = {Braun, Gábor and Pokutta, Sebastian}, title = {Dual Prices for Frank--Wolfe Algorithms}, date = {2021-01-06} } - Roux, C., Wirth, E., Pokutta, S., and Kerdreux, T. (2021). Efficient Online-Bandit Strategies for Minimax Learning Problems.
[arXiv]
[BibTeX]
@misc{2021_RouxWirthPokuttaKerdreux_Onlinebanditminimax, archiveprefix = {arXiv}, eprint = {2105.13939}, arxiv = {arXiv:2105.13939}, primaryclass = {cs.LG}, author = {Roux, Christophe and Wirth, Elias and Pokutta, Sebastian and Kerdreux, Thomas}, title = {Efficient Online-Bandit Strategies for Minimax Learning Problems}, date = {2021-05-28} } - Braun, G., Pokutta, S., and Weismantel, R. (2022). Alternating Linear Minimization: Revisiting von Neumann’s alternating projections.
[arXiv]
[slides]
[video]
[BibTeX]
@misc{2022_BraunPokuttaWeismantel_Alternatingminimization, archiveprefix = {arXiv}, eprint = {2212.02933}, arxiv = {arXiv:2212.02933}, primaryclass = {math.OC}, author = {Braun, Gábor and Pokutta, Sebastian and Weismantel, Robert}, title = {Alternating Linear Minimization: Revisiting von Neumann's alternating projections}, slides = {https://pokutta.com/slides/20230327-icerm.pdf}, video = {https://icerm.brown.edu/programs/sp-s23/w2/#schedule-item-4945}, date = {2022-12-06} } - Scieur, D., Martínez-Rubio, D., Kerdreux, T., d’Aspremont, A., and Pokutta, S. (2023). Strong Convexity of Sets in Riemannian Manifolds.
[arXiv]
[BibTeX]
@misc{2023_KerdreuxScieurDaspremontPokutta_StrongconvexityRiemannianmanifolds, archiveprefix = {arXiv}, eprint = {2312.03583}, arxiv = {arXiv:2312.03583}, primaryclass = {math.OC}, author = {Scieur, Damien and Martínez-Rubio, David and Kerdreux, Thomas and d'Aspremont, Alexandre and Pokutta, Sebastian}, title = {Strong Convexity of Sets in Riemannian Manifolds}, date = {2023-12-06} } - Aigner, K.-M., Denzler, S., Liers, F., Pokutta, S., and Sharma, K. (2025). Scenario Reduction for Distributionally Robust Optimization.
[arXiv]
[BibTeX]
@misc{2025_Kevin-martinEtAl_ScenarioreductionDro, archiveprefix = {arXiv}, eprint = {2503.11484}, arxiv = {arXiv:2503.11484}, primaryclass = {math.OC}, author = {Aigner, Kevin-Martin and Denzler, Sebastian and Liers, Frauke and Pokutta, Sebastian and Sharma, Kartikey}, title = {Scenario Reduction for Distributionally Robust Optimization}, date = {2025-03-14} } - Pokutta, S. (2025). Scalable DC Optimization via Adaptive Frank-Wolfe Algorithms.
[arXiv]
[BibTeX]
@misc{2025_Pokutta_DcoptimizationFrankwolfe_2507-17545, archiveprefix = {arXiv}, eprint = {2507.17545}, arxiv = {arXiv:2507.17545}, primaryclass = {math.OC}, author = {Pokutta, Sebastian}, title = {Scalable DC Optimization via Adaptive Frank-Wolfe Algorithms}, date = {2025-07-23} } - Takahashi, S., Pokutta, S., and Takeda, A. (2025). Fast Frank–Wolfe Algorithms with Adaptive Bregman Step-Size for Weakly Convex Functions.
[arXiv]
[BibTeX]
@misc{2025_ShotaPokuttaAkiko_Frankwolfebregman, archiveprefix = {arXiv}, eprint = {2504.04330}, arxiv = {arXiv:2504.04330}, primaryclass = {math.OC}, author = {Takahashi, Shota and Pokutta, Sebastian and Takeda, Akiko}, title = {Fast Frank--Wolfe Algorithms with Adaptive Bregman Step-Size for Weakly Convex Functions}, date = {2025-04-06} } - Zimmer, M., Roux, C., Wagner, M., Hendrych, D., and Pokutta, S. (2025). SparseSwaps: Tractable LLM Pruning Mask Refinement at Scale.
[arXiv]
[BibTeX]
@misc{2025_ZimmerEtAl_Sparseswapspruning_2512-10922, archiveprefix = {arXiv}, eprint = {2512.10922}, arxiv = {arXiv:2512.10922}, primaryclass = {cs.LG}, author = {Zimmer, Max and Roux, Christophe and Wagner, Moritz and Hendrych, Deborah and Pokutta, Sebastian}, title = {SparseSwaps: Tractable LLM Pruning Mask Refinement at Scale}, date = {2025-12-11} } - Elhishi, A., Roux, C., and d’Aspremont, A. (2026). Restart and Adaptive Acceleration in Stochastic Gradient Methods.
[arXiv]
[BibTeX]
@misc{2026_ElhishiRouxD'aspremont_Restartacceleration_2606-21354, archiveprefix = {arXiv}, eprint = {2606.21354}, arxiv = {arXiv:2606.21354}, primaryclass = {math.OC}, author = {Elhishi, Ali and Roux, Christophe and d'Aspremont, Alexandre}, title = {Restart and Adaptive Acceleration in Stochastic Gradient Methods}, date = {2026-06-19} } - Pokutta, S. (2026). Lower Bounds for Randomized Linear-Oracle Optimization over Strongly Convex Sets.
[arXiv]
[BibTeX]
@misc{2026_Pokutta_RandomizedLinearOracle_2610-13590, archiveprefix = {arXiv}, eprint = {2610.13590}, arxiv = {arXiv:2610.13590}, primaryclass = {math.OC}, author = {Pokutta, Sebastian}, title = {Lower Bounds for Randomized Linear-Oracle Optimization over Strongly Convex Sets}, date = {2026-10-09} } - Schiekiera, L., Zimmer, M., Roux, C., Arnold, M., Pokutta, S., and Günther, F. (2026). LinearPFN: Amortized Variable Selection for Linear Models with Interactions.
[arXiv]
[BibTeX]
@misc{2026_SchiekieraEtAl_LinearPFNVariableSelection_2610-06580, archiveprefix = {arXiv}, eprint = {2610.06580}, arxiv = {arXiv:2610.06580}, primaryclass = {cs.LG}, author = {Schiekiera, Louis and Zimmer, Max and Roux, Christophe and Arnold, Manuel and Pokutta, Sebastian and Günther, Fritz}, title = {LinearPFN: Amortized Variable Selection for Linear Models with Interactions}, date = {2026-10-05} } - Wagner, M., Roux, C., Zimmer, M., and Pokutta, S. (2026). The KV-Surgeon: Compensation-Aware Offline KV-Cache Compaction.
[arXiv]
[BibTeX]
@misc{2026_WagnerRouxZimmerPokutta_KVCacheCompaction_2610-25933, archiveprefix = {arXiv}, eprint = {2610.25933}, arxiv = {arXiv:2610.25933}, primaryclass = {cs.LG}, author = {Wagner, Moritz and Roux, Christophe and Zimmer, Max and Pokutta, Sebastian}, title = {The KV-Surgeon: Compensation-Aware Offline KV-Cache Compaction}, date = {2026-10-19} } - Xu, L., Zhou, Y., and Pokutta, S. (2026). Agentic MIP Research: Accelerated Constraint Handler Generation.
[arXiv]
[BibTeX]
@misc{2026_XuZhouPokutta_Agenticmip_2605-09186, archiveprefix = {arXiv}, eprint = {2605.09186}, arxiv = {arXiv:2605.09186}, primaryclass = {cs.AI}, author = {Xu, Liding and Zhou, Yugeng and Pokutta, Sebastian}, title = {Agentic MIP Research: Accelerated Constraint Handler Generation}, date = {2026-05-09} } - Zhou, Z., Li, Z., Huang, W., Li, X., Cao, C., Feng, X., Lu, X., Hu, J., Xie, Y., Pelleriti, N., Liu, S., Lu, M., Zimmer, M., Miranda, B., Yao, J., Liu, B., Koyejo, S., Pokutta, S., and Han, B. (2026). AlphaDiana: A Measurement System for Agents in Verifiable Reasoning Tasks.
[arXiv]
[BibTeX]
@misc{2026_ZhouEtAl_AlphaDianaVerifiableReasoning_2610-20675, archiveprefix = {arXiv}, eprint = {2610.20675}, arxiv = {arXiv:2610.20675}, primaryclass = {cs.AI}, author = {Zhou, Zhanke and Li, Zongze and Huang, Weikai and Li, Xuan and Cao, Chentao and Feng, Xiao and Lu, Xiangyu and Hu, Jinbo and Xie, Yi and Pelleriti, Nico and Liu, Shiyang and Lu, Menghan and Zimmer, Max and Miranda, Brando and Yao, Jiangchao and Liu, Bo and Koyejo, Sanmi and Pokutta, Sebastian and Han, Bo}, title = {AlphaDiana: A Measurement System for Agents in Verifiable Reasoning Tasks}, date = {2026-10-14} } - Zimmer, M., Pelleriti, N., Roux, C., and Pokutta, S. (2026). The Agentic Researcher: A Practical Guide to AI-Assisted Research in Mathematics and Machine Learning.
[arXiv]
[BibTeX]
@misc{2026_ZimmerPelleritiRouxPokutta_Airesearchframework_2603-15914, archiveprefix = {arXiv}, eprint = {2603.15914}, arxiv = {arXiv:2603.15914}, primaryclass = {cs.LG}, author = {Zimmer, Max and Pelleriti, Nico and Roux, Christophe and Pokutta, Sebastian}, title = {The Agentic Researcher: A Practical Guide to AI-Assisted Research in Mathematics and Machine Learning}, date = {2026-03-16} }
Conference contributions
- Halbey, J., Roux, C., and Pokutta, S. Curvature-Dependent Lower Bounds for Frank-Wolfe. Advances in Neural Information Processing Systems, 39.
To appear.
Associated with NeurIPS 2026.
[arXiv]
[BibTeX]
@inproceedings{2026_HalbeyRouxPokutta_Curvaturelowerbounds_2605-10595, booktitle = {Advances in Neural Information Processing Systems}, series = {Advances in Neural Information Processing Systems}, volume = {39}, publisher = {Curran Associates}, note = {To appear}, archiveprefix = {arXiv}, eprint = {2605.10595}, arxiv = {arXiv:2605.10595}, primaryclass = {math.OC}, author = {Halbey, Jannis and Roux, Christophe and Pokutta, Sebastian}, title = {Curvature-Dependent Lower Bounds for Frank-Wolfe} } - Koshizuka, T., Blessing, D., Zimmer, M., Pokutta, S., and Richter, L. A Trust Region Approach for Learning Schrödinger Bridges. Advances in Neural Information Processing Systems, 39.
To appear.
Associated with NeurIPS 2026.
[BibTeX]
@inproceedings{2026_KoshizukaEtAl_TrustregionSchrodingerbridges, booktitle = {Advances in Neural Information Processing Systems}, series = {Advances in Neural Information Processing Systems}, volume = {39}, publisher = {Curran Associates}, note = {To appear}, author = {Koshizuka, Takeshi and Blessing, Denis and Zimmer, Max and Pokutta, Sebastian and Richter, Lorenz}, title = {A Trust Region Approach for Learning Schrödinger Bridges} } - Pokutta, S. Frank-Wolfe Beyond 1/t Convergence. Advances in Neural Information Processing Systems, 39.
To appear.
Associated with NeurIPS 2026.
[arXiv]
[BibTeX]
@inproceedings{2026_Pokutta_Localdualsharpness_2604-28006, booktitle = {Advances in Neural Information Processing Systems}, series = {Advances in Neural Information Processing Systems}, volume = {39}, publisher = {Curran Associates}, note = {To appear}, archiveprefix = {arXiv}, eprint = {2604.28006}, arxiv = {arXiv:2604.28006}, primaryclass = {math.OC}, author = {Pokutta, Sebastian}, title = {Frank-Wolfe Beyond 1/t Convergence} } - Halbey, J., Deza, D., Zimmer, M., Roux, C., Stellato, B., and Pokutta, S. (2026). Lower Bounds for Frank-Wolfe on Strongly Convex Sets. Proceedings of the 43rd International Conference on Machine Learning, 306.
Associated with ICML 2026.
[arXiv]
[BibTeX]
@inproceedings{2026_HalbeyEtAl_Lowerboundsfrankwolfe_2602-04378, year = {2026}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, series = {Proceedings of Machine Learning Research}, volume = {306}, month = jul, archiveprefix = {arXiv}, eprint = {2602.04378}, arxiv = {arXiv:2602.04378}, primaryclass = {math.OC}, author = {Halbey, Jannis and Deza, Daniel and Zimmer, Max and Roux, Christophe and Stellato, Bartolomeo and Pokutta, Sebastian}, title = {Lower Bounds for Frank-Wolfe on Strongly Convex Sets} } - Schiekiera, L., Zimmer, M., Roux, C., Pokutta, S., and Günther, F. (2026). From Associations to Activations: Comparing Behavioral and Hidden-State Semantic Geometry in LLMs. Proceedings of the 43rd International Conference on Machine Learning, 306.
Associated with ICML 2026.
[arXiv]
[BibTeX]
@inproceedings{2026_LouisEtAl_Behavioralgeometry_2602-00628, year = {2026}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, series = {Proceedings of Machine Learning Research}, volume = {306}, month = jul, archiveprefix = {arXiv}, eprint = {2602.00628}, arxiv = {arXiv:2602.00628}, primaryclass = {cs.LG}, author = {Schiekiera, Louis and Zimmer, Max and Roux, Christophe and Pokutta, Sebastian and Günther, Fritz}, title = {From Associations to Activations: Comparing Behavioral and Hidden-State Semantic Geometry in LLMs} } - Iommazzo, G., Martínez-Rubio, D., Criado, F., Wirth, E., and Pokutta, S. (2026). Linear Convergence of the Frank-Wolfe Algorithm over Product Polytopes. Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, 300.
Associated with AISTATS 2026.
[arXiv]
[BibTeX]
@inproceedings{2025_IommazzoEtAl_Frankwolfeconvergence, year = {2026}, booktitle = {Proceedings of The 29th International Conference on Artificial Intelligence and Statistics}, series = {Proceedings of Machine Learning Research}, volume = {300}, archiveprefix = {arXiv}, eprint = {2505.11259}, arxiv = {arXiv:2505.11259}, primaryclass = {math.OC}, author = {Iommazzo, Gabriele and Martínez-Rubio, David and Criado, Francisco and Wirth, Elias and Pokutta, Sebastian}, title = {Linear Convergence of the Frank-Wolfe Algorithm over Product Polytopes} } - Xiao, W., Besançon, M., Gelß, P., Hendrych, D., Klus, S., and Pokutta, S. (2026). Graph Isomorphism: Mixed-Integer Convex Optimization from First-Order Methods. Integration of Constraint Programming, Artificial Intelligence, and Operations Research, 16595, 614–630.
Associated with CPAIOR 2026.
DOI: 10.1007/978-3-032-27242-3_36
[arXiv]
[BibTeX]
@inproceedings{2025_XiaoEtAl_GraphisomorphismOptimization_2512-17417, year = {2026}, booktitle = {Integration of Constraint Programming, Artificial Intelligence, and Operations Research}, series = {Lecture Notes in Computer Science}, volume = {16595}, publisher = {Springer Science+Business Media}, pages = {614--630}, doi = {10.1007/978-3-032-27242-3_36}, archiveprefix = {arXiv}, eprint = {2512.17417}, arxiv = {arXiv:2512.17417}, primaryclass = {math.OC}, author = {Xiao, Wenjie and Besançon, Mathieu and Gelß, Patrick and Hendrych, Deborah and Klus, Stefan and Pokutta, Sebastian}, title = {Graph Isomorphism: Mixed-Integer Convex Optimization from First-Order Methods} } - Pelleriti, N., Kraus, N., May, A., and Pokutta, S. (2026). CryptHarness: A Harness for Agent-Assisted Cryptanalytic Research. Accepted at NeurIPS 2026 Workshop on SaTQuML: Secure and Trustworthy Quantum Machine Learning.
[BibTeX]
@misc{2026_PelleritiKrausMayPokutta_CryptHarness, howpublished = {Accepted at NeurIPS 2026 Workshop on SaTQuML: Secure and Trustworthy Quantum Machine Learning}, year = {2026}, author = {Pelleriti, Nico and Kraus, Nicolai and May, Alexander and Pokutta, Sebastian}, title = {CryptHarness: A Harness for Agent-Assisted Cryptanalytic Research} } - Pelleriti, N., Kraus, N., May, A., and Pokutta, S. (2026). CryptHarness: A Harness for Agent-Assisted Cryptanalytic Research. Accepted at Third Workshop on Agents in the Wild: Safety, Security, and Beyond @ NeurIPS 2026.
[BibTeX]
@misc{2026_PelleritiKrausMayPokutta_CryptHarness:1, howpublished = {Accepted at Third Workshop on Agents in the Wild: Safety, Security, and Beyond @ NeurIPS 2026}, year = {2026}, author = {Pelleriti, Nico and Kraus, Nicolai and May, Alexander and Pokutta, Sebastian}, title = {CryptHarness: A Harness for Agent-Assisted Cryptanalytic Research} } - Pelleriti, N., Mundinger, K., Zimmer, M., and Pokutta, S. (2026). Zero: A Meta-Agent for Mathematical Research. Accepted at The Sixth Workshop on Mathematical Reasoning and AI @ NeurIPS 2026.
[BibTeX]
@misc{2026_PelleritiMundingerZimmerPokutta_Zero, howpublished = {Accepted at The Sixth Workshop on Mathematical Reasoning and AI @ NeurIPS 2026}, year = {2026}, author = {Pelleriti, Nico and Mundinger, Konrad and Zimmer, Max and Pokutta, Sebastian}, title = {Zero: A Meta-Agent for Mathematical Research} } - Pelleriti, N., Mundinger, K., Zimmer, M., and Pokutta, S. (2026). Zero: A Meta-Agent for Mathematical Research. Accepted at First Workshop on Meta Agents: Managing Agents That Manage Agents @ NeurIPS 2026.
[BibTeX]
@misc{2026_PelleritiMundingerZimmerPokutta_Zero:1, howpublished = {Accepted at First Workshop on Meta Agents: Managing Agents That Manage Agents @ NeurIPS 2026}, year = {2026}, author = {Pelleriti, Nico and Mundinger, Konrad and Zimmer, Max and Pokutta, Sebastian}, title = {Zero: A Meta-Agent for Mathematical Research} } - Petit-Tichanné, U., Pelleriti, N., Zimmer, M., and Pokutta, S. (2026). Learning the Coefficients of the Standard Volatility Forecaster. Accepted at NeurIPS 2026 Workshop TS-LIMITS: Generalization for Time Series in Tight Settings: Latency, Inference, Memory, PrIvacy and SusTainability.
[BibTeX]
@misc{2026_PetitTichannePelleritiZimmerPokutta_Volatilityforecaster, howpublished = {Accepted at NeurIPS 2026 Workshop TS-LIMITS: Generalization for Time Series in Tight Settings: Latency, Inference, Memory, PrIvacy and SusTainability}, year = {2026}, author = {Petit-Tichanné, Ulysse and Pelleriti, Nico and Zimmer, Max and Pokutta, Sebastian}, title = {Learning the Coefficients of the Standard Volatility Forecaster} } - Zhou, Z., Li, Z., Huang, W., Li, X., Cao, C., Feng, X., Lu, X., Hu, J., Lu, M., Xie, Y., Pelleriti, N., Liu, S., Zimmer, M., Miranda, B., Yao, J., Liu, B., Koyejo, S., Pokutta, S., and Han, B. (2026). Reasoning Is More Than the Model: Harness-Aware Evaluation of Agents on Verifiable Reasoning Tasks. Accepted at ICML 2026 Workshop on AI As a Tool for Mathematics, Computer Science, and Machine Learning.
[arXiv]
[BibTeX]
@misc{2026_ZhouEtAl_AlphaDianaVerifiableReasoning_2610-20675:1, howpublished = {Accepted at ICML 2026 Workshop on AI As a Tool for Mathematics, Computer Science, and Machine Learning}, year = {2026}, archiveprefix = {arXiv}, eprint = {2610.20675}, arxiv = {arXiv:2610.20675}, primaryclass = {cs.AI}, author = {Zhou, Zhanke and Li, Zongze and Huang, Weikai and Li, Xuan and Cao, Chentao and Feng, Xiao and Lu, Xiangyu and Hu, Jinbo and Lu, Menghan and Xie, Yi and Pelleriti, Nico and Liu, Shiyang and Zimmer, Max and Miranda, Brando and Yao, Jiangchao and Liu, Bo and Koyejo, Sanmi and Pokutta, Sebastian and Han, Bo}, title = {Reasoning Is More Than the Model: Harness-Aware Evaluation of Agents on Verifiable Reasoning Tasks} } - Zhou, Z., Li, Z., Huang, W., Li, X., Cao, C., Feng, X., Lu, X., Hu, J., Lu, M., Xie, Y., Pelleriti, N., Liu, S., Zimmer, M., Miranda, B., Yao, J., Liu, B., Koyejo, S., Pokutta, S., and Han, B. (2026). Reasoning Is More Than the Model: Harness-Aware Evaluation of Agents on Verifiable Reasoning Tasks. Accepted at 3rd AI for Math Workshop: Toward Self-Evolving Scientific Agents @ ICML 2026.
[arXiv]
[BibTeX]
@misc{2026_ZhouEtAl_AlphaDianaVerifiableReasoning_2610-20675:2, howpublished = {Accepted at 3rd AI for Math Workshop: Toward Self-Evolving Scientific Agents @ ICML 2026}, year = {2026}, archiveprefix = {arXiv}, eprint = {2610.20675}, arxiv = {arXiv:2610.20675}, primaryclass = {cs.AI}, author = {Zhou, Zhanke and Li, Zongze and Huang, Weikai and Li, Xuan and Cao, Chentao and Feng, Xiao and Lu, Xiangyu and Hu, Jinbo and Lu, Menghan and Xie, Yi and Pelleriti, Nico and Liu, Shiyang and Zimmer, Max and Miranda, Brando and Yao, Jiangchao and Liu, Bo and Koyejo, Sanmi and Pokutta, Sebastian and Han, Bo}, title = {Reasoning Is More Than the Model: Harness-Aware Evaluation of Agents on Verifiable Reasoning Tasks} } - Zhou, Z., Li, Z., Huang, W., Li, X., Cao, C., Feng, X., Lu, X., Hu, J., Lu, M., Xie, Y., Pelleriti, N., Liu, S., Zimmer, M., Miranda, B., Yao, J., Liu, B., Koyejo, S., Pokutta, S., and Han, B. (2026). AlphaDiana: A System for Evaluating Reasoning Agents. Accepted at NeurIPS 2026 Workshop on Towards Test-Time Continual Learning Agents.
[arXiv]
[BibTeX]
@misc{2026_ZhouEtAl_AlphaDianaVerifiableReasoning_2610-20675:3, howpublished = {Accepted at NeurIPS 2026 Workshop on Towards Test-Time Continual Learning Agents}, year = {2026}, archiveprefix = {arXiv}, eprint = {2610.20675}, arxiv = {arXiv:2610.20675}, primaryclass = {cs.AI}, author = {Zhou, Zhanke and Li, Zongze and Huang, Weikai and Li, Xuan and Cao, Chentao and Feng, Xiao and Lu, Xiangyu and Hu, Jinbo and Lu, Menghan and Xie, Yi and Pelleriti, Nico and Liu, Shiyang and Zimmer, Max and Miranda, Brando and Yao, Jiangchao and Liu, Bo and Koyejo, Sanmi and Pokutta, Sebastian and Han, Bo}, title = {AlphaDiana: A System for Evaluating Reasoning Agents} } - Zhou, Z., Li, Z., Huang, W., Li, X., Cao, C., Feng, X., Lu, X., Hu, J., Lu, M., Xie, Y., Pelleriti, N., Liu, S., Zimmer, M., Miranda, B., Yao, J., Liu, B., Koyejo, S., Pokutta, S., and Han, B. (2026). AlphaDiana: A System for Evaluating Reasoning Agents. Accepted at The Sixth Workshop on Mathematical Reasoning and AI @ NeurIPS 2026.
[arXiv]
[BibTeX]
@misc{2026_ZhouEtAl_AlphaDianaVerifiableReasoning_2610-20675:4, howpublished = {Accepted at The Sixth Workshop on Mathematical Reasoning and AI @ NeurIPS 2026}, year = {2026}, archiveprefix = {arXiv}, eprint = {2610.20675}, arxiv = {arXiv:2610.20675}, primaryclass = {cs.AI}, author = {Zhou, Zhanke and Li, Zongze and Huang, Weikai and Li, Xuan and Cao, Chentao and Feng, Xiao and Lu, Xiangyu and Hu, Jinbo and Lu, Menghan and Xie, Yi and Pelleriti, Nico and Liu, Shiyang and Zimmer, Max and Miranda, Brando and Yao, Jiangchao and Liu, Bo and Koyejo, Sanmi and Pokutta, Sebastian and Han, Bo}, title = {AlphaDiana: A System for Evaluating Reasoning Agents} } - Halbey, J., Rakotomandimby, S., Besançon, M., Designolle, S., and Pokutta, S. (2025). Efficient Quadratic Corrections for Frank-Wolfe Algorithms. Advances in Neural Information Processing Systems, 38.
Associated with NeurIPS 2025.
[arXiv]
[BibTeX]
@inproceedings{2025_HalbeyRakotomandimbyBesanconDesignollePokutta_Quadraticcorrections_2506-02635, year = {2025}, booktitle = {Advances in Neural Information Processing Systems}, series = {Advances in Neural Information Processing Systems}, volume = {38}, publisher = {Curran Associates}, month = sep, archiveprefix = {arXiv}, eprint = {2506.02635}, arxiv = {arXiv:2506.02635}, primaryclass = {math.OC}, author = {Halbey, Jannis and Rakotomandimby, Seta and Besançon, Mathieu and Designolle, Sébastien and Pokutta, Sebastian}, title = {Efficient Quadratic Corrections for Frank-Wolfe Algorithms} } - Hendrych, D., Besançon, M., Martínez-Rubio, D., and Pokutta, S. (2025). Secant Line Search for Frank-Wolfe Algorithms. Proceedings of the 42nd International Conference on Machine Learning, 267.
Associated with ICML 2025.
[arXiv]
[BibTeX]
@inproceedings{2025_HendrychBesanconMartinezrubioPokutta_Secantfrankwolfe, year = {2025}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, series = {Proceedings of Machine Learning Research}, volume = {267}, month = may, archiveprefix = {arXiv}, eprint = {2501.18775}, arxiv = {arXiv:2501.18775}, primaryclass = {math.OC}, author = {Hendrych, Deborah and Besançon, Mathieu and Martínez-Rubio, David and Pokutta, Sebastian}, title = {Secant Line Search for Frank-Wolfe Algorithms} } - Pelleriti, N., Zimmer, M., Wirth, E., and Pokutta, S. (2025). Approximating Latent Manifolds in Neural Networks via Vanishing Ideals. Proceedings of the 42nd International Conference on Machine Learning, 267.
Associated with ICML 2025.
[arXiv]
[BibTeX]
@inproceedings{2025_PelleritiZimmerWirthPokutta_Latentmanifolds, year = {2025}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, series = {Proceedings of Machine Learning Research}, volume = {267}, month = may, archiveprefix = {arXiv}, eprint = {2502.15051}, arxiv = {arXiv:2502.15051}, primaryclass = {cs.LG}, author = {Pelleriti, Nico and Zimmer, Max and Wirth, Elias and Pokutta, Sebastian}, title = {Approximating Latent Manifolds in Neural Networks via Vanishing Ideals} } - Roux, C., Martínez-Rubio, D., and Pokutta, S. (2025). Implicit Riemannian Optimism with Applications to Min-Max Problems. Proceedings of the 42nd International Conference on Machine Learning, 267, 52139–52172.
Associated with ICML 2025.
[arXiv]
[BibTeX]
@inproceedings{2025_RouxMartinezrubioPokutta_ImplicitRiemannian, year = {2025}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, series = {Proceedings of Machine Learning Research}, volume = {267}, month = may, pages = {52139--52172}, archiveprefix = {arXiv}, eprint = {2501.18381}, arxiv = {arXiv:2501.18381}, primaryclass = {math.OC}, author = {Roux, Christophe and Martínez-Rubio, David and Pokutta, Sebastian}, title = {Implicit Riemannian Optimism with Applications to Min-Max Problems} } - Martínez-Rubio, D., Roux, C., Criscitiello, C., and Pokutta, S. (2025). Accelerated Methods for Riemannian Min-Max Optimization Ensuring Bounded Geometric Penalties. Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, 258, 280–288.
Associated with AISTATS 2025.
[arXiv]
[BibTeX]
@inproceedings{2023_MartinezrubioRouxCriscitielloPokutta_Riemannianminmax, year = {2025}, booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics}, series = {Proceedings of Machine Learning Research}, volume = {258}, month = jan, pages = {280--288}, archiveprefix = {arXiv}, eprint = {2305.16186}, arxiv = {arXiv:2305.16186}, primaryclass = {math.OC}, author = {Martínez-Rubio, David and Roux, Christophe and Criscitiello, Christopher and Pokutta, Sebastian}, title = {Accelerated Methods for Riemannian Min-Max Optimization Ensuring Bounded Geometric Penalties} } - Wirth, E., Besançon, M., and Pokutta, S. (2025). The Pivoting Framework: Frank-Wolfe Algorithms with Active Set Size Control. Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, 258, 271–279.
Associated with AISTATS 2025.
[URL]
[arXiv]
[BibTeX]
@inproceedings{2024_WirthBesanconPokutta_Pivotingframework, year = {2025}, booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics}, series = {Proceedings of Machine Learning Research}, volume = {258}, month = jan, pages = {271--279}, url = {https://proceedings.mlr.press/v258/besancon25a.html}, archiveprefix = {arXiv}, eprint = {2407.11760}, arxiv = {arXiv:2407.11760}, primaryclass = {math.OC}, author = {Wirth, Elias and Besançon, Mathieu and Pokutta, Sebastian}, title = {The Pivoting Framework: Frank-Wolfe Algorithms with Active Set Size Control} } - Martínez-Rubio, D., Roux, C., and Pokutta, S. (2024). Convergence and Trade-Offs in Riemannian Gradient Descent and Riemannian Proximal Point. Proceedings of the 41st International Conference on Machine Learning, 235.
Associated with ICML 2024.
[URL]
[arXiv]
[BibTeX]
@inproceedings{2024_MartinezrubioRouxPokutta_Riemanniangradientdescent, year = {2024}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, series = {Proceedings of Machine Learning Research}, volume = {235}, url = {https://proceedings.mlr.press/v235/marti-nez-rubio24a.html}, archiveprefix = {arXiv}, eprint = {2403.10429}, arxiv = {arXiv:2403.10429}, primaryclass = {math.OC}, author = {Martínez-Rubio, David and Roux, Christophe and Pokutta, Sebastian}, title = {Convergence and Trade-Offs in Riemannian Gradient Descent and Riemannian Proximal Point} } - Martínez-Rubio, D., and Pokutta, S. (2023). Accelerated Riemannian Optimization: Handling Constraints with a Prox to Bound Geometric Penalties. Proceedings of Thirty Sixth Conference on Learning Theory, 195.
Associated with COLT 2023.
[URL]
[arXiv]
[poster]
[BibTeX]
@inproceedings{2022_MartinezrubioPokutta_Acceleratedriemannian, year = {2023}, booktitle = {Proceedings of Thirty Sixth Conference on Learning Theory}, series = {Proceedings of Machine Learning Research}, volume = {195}, month = may, url = {https://proceedings.mlr.press/v195/martinez-rubio23a/martinez-rubio23a.pdf}, archiveprefix = {arXiv}, eprint = {2211.14645}, arxiv = {arXiv:2211.14645}, primaryclass = {math.OC}, author = {Martínez-Rubio, David and Pokutta, Sebastian}, title = {Accelerated Riemannian Optimization: Handling Constraints with a Prox to Bound Geometric Penalties}, poster = {https://pokutta.com/slides/20221203_poster_neurips_riemannian.pdf} } - Martínez-Rubio, D., Roux, C., Criscitiello, C., and Pokutta, S. (2023). Accelerated Methods for Riemannian Min-Max Optimization Ensuring Bounded Geometric Penalties. Accepted at Optimization for Machine Learning.
[arXiv]
[BibTeX]
@misc{2023_MartinezrubioRouxCriscitielloPokutta_Riemannianminmax:1, howpublished = {Accepted at Optimization for Machine Learning}, year = {2023}, month = may, archiveprefix = {arXiv}, eprint = {2305.16186}, arxiv = {arXiv:2305.16186}, primaryclass = {math.OC}, author = {Martínez-Rubio, David and Roux, Christophe and Criscitiello, Christopher and Pokutta, Sebastian}, title = {Accelerated Methods for Riemannian Min-Max Optimization Ensuring Bounded Geometric Penalties} } - Martínez-Rubio, D., Wirth, E., and Pokutta, S. (2023). Accelerated and Sparse Algorithms for Approximate Personalized PageRank and Beyond. Proceedings of Thirty Sixth Conference on Learning Theory, 195.
Associated with COLT 2023.
[arXiv]
[BibTeX]
@inproceedings{2023_MartinezrubioWirthPokutta_Sparseapproximatepagerank, year = {2023}, booktitle = {Proceedings of Thirty Sixth Conference on Learning Theory}, series = {Proceedings of Machine Learning Research}, volume = {195}, month = may, archiveprefix = {arXiv}, eprint = {2303.12875}, arxiv = {arXiv:2303.12875}, primaryclass = {math.OC}, author = {Martínez-Rubio, David and Wirth, Elias and Pokutta, Sebastian}, title = {Accelerated and Sparse Algorithms for Approximate Personalized PageRank and Beyond} } - Wirth, E., Kera, H., and Pokutta, S. (2023, January). Approximate Vanishing Ideal Computations at Scale. 11th International Conference on Learning Representations (ICLR 2023).
Associated with ICLR 2023.
[arXiv]
[slides]
[BibTeX]
@inproceedings{2022_WirthKeraPokutta_Approximatevanishingideal, year = {2023}, booktitle = {11th International Conference on Learning Representations (ICLR 2023)}, publisher = {ICLR (OpenReview)}, month = jan, archiveprefix = {arXiv}, eprint = {2207.01236}, arxiv = {arXiv:2207.01236}, primaryclass = {cs.LG}, author = {Wirth, Elias and Kera, Hiroshi and Pokutta, Sebastian}, title = {Approximate Vanishing Ideal Computations at Scale}, slides = {https://pokutta.com/slides/20220915_avi_at_scale.pdf} } - Wirth, E., Kerdreux, T., and Pokutta, S. (2023). Acceleration of Frank-Wolfe Algorithms with Open-Loop Step-Sizes. Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, 206, 77–100.
Associated with AISTATS 2023.
[URL]
[arXiv]
[BibTeX]
@inproceedings{2022_WirthKerdreuxPokutta_Frankwolfeacceleration, year = {2023}, booktitle = {Proceedings of The 26th International Conference on Artificial Intelligence and Statistics}, series = {Proceedings of Machine Learning Research}, volume = {206}, month = jan, pages = {77--100}, url = {https://proceedings.mlr.press/v206/wirth23a.html}, archiveprefix = {arXiv}, eprint = {2205.12838}, arxiv = {arXiv:2205.12838}, primaryclass = {math.OC}, author = {Wirth, Elias and Kerdreux, Thomas and Pokutta, Sebastian}, title = {Acceleration of Frank-Wolfe Algorithms with Open-Loop Step-Sizes} } - Criscitiello, C., Martínez-Rubio, D., and Boumal, N. (2023). Open Problem: Polynomial linearly-convergent method for g-convex optimization? Proceedings of Thirty Sixth Conference on Learning Theory, 195.
Associated with COLT 2023.
[arXiv]
[BibTeX]
@inproceedings{2023_CriscitielloMartinezrubioBoumal_Polynomialconvergence, year = {2023}, booktitle = {Proceedings of Thirty Sixth Conference on Learning Theory}, series = {Proceedings of Machine Learning Research}, volume = {195}, archiveprefix = {arXiv}, eprint = {2307.12743}, arxiv = {arXiv:2307.12743}, primaryclass = {math.OC}, author = {Criscitiello, Christopher and Martínez-Rubio, David and Boumal, Nicolas}, title = {Open Problem: Polynomial linearly-convergent method for g-convex optimization?} } - Macdonald, J., Besançon, M., and Pokutta, S. (2022). Interpretable Neural Networks with Frank-Wolfe: Sparse Relevance Maps and Relevance Orderings. Proceedings of the 39th International Conference on Machine Learning, 162.
Associated with ICML 2022.
[arXiv]
[poster]
[video]
[BibTeX]
@inproceedings{2021_MacdonaldBesanconPokutta_Interpretableneuralnetworks, year = {2022}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, series = {Proceedings of Machine Learning Research}, volume = {162}, month = may, archiveprefix = {arXiv}, eprint = {2110.08105}, arxiv = {arXiv:2110.08105}, primaryclass = {cs.LG}, author = {Macdonald, Jan and Besançon, Mathieu and Pokutta, Sebastian}, title = {Interpretable Neural Networks with Frank-Wolfe: Sparse Relevance Maps and Relevance Orderings}, poster = {https://pokutta.com/slides/20220712_icml_poster_interpretable_rde.pdf}, video = {https://slideslive.com/38983588} } - Búi, M. N., Combettes, P. L., and Woodstock, Z. (2022). Block-Activated Algorithms For Multicomponent Fully Nonsmooth Minimization. ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 5428–5432.
Associated with ICASSP 2022.
DOI: 10.1109/ICASSP43922.2022.9747479
[URL]
[arXiv]
[BibTeX]
@inproceedings{2022_BiCombettesWoodstock_Blockactivatedminimization, year = {2022}, booktitle = {ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, publisher = {Institute of Electrical and Electronics Engineers}, month = may, pages = {5428--5432}, doi = {10.1109/ICASSP43922.2022.9747479}, url = {https://zevwoodstock.github.io/media/publications/icassp2022-2.pdf}, archiveprefix = {arXiv}, eprint = {2103.00520}, arxiv = {arXiv:2103.00520}, primaryclass = {math.OC}, author = {Búi, M. N. and Combettes, Patrick L. and Woodstock, Zev}, title = {Block-Activated Algorithms For Multicomponent Fully Nonsmooth Minimization} } - Wirth, E., and Pokutta, S. (2022). Conditional Gradients for the Approximately Vanishing Ideal. Proceedings of The 25th International Conference on Artificial Intelligence and Statistics, 151.
Associated with AISTATS 2022.
[arXiv]
[summary]
[poster]
[code]
[BibTeX]
@inproceedings{2022_WirthPokutta_Conditionalgradients, year = {2022}, booktitle = {Proceedings of The 25th International Conference on Artificial Intelligence and Statistics}, series = {Proceedings of Machine Learning Research}, volume = {151}, month = jan, archiveprefix = {arXiv}, eprint = {2202.03349}, arxiv = {arXiv:2202.03349}, primaryclass = {cs.LG}, author = {Wirth, Elias and Pokutta, Sebastian}, title = {Conditional Gradients for the Approximately Vanishing Ideal}, code = {https://github.com/ZIB-IOL/cgavi/}, poster = {https://pokutta.com/slides/20220223_CGAVI_poster.pdf}, summary = {https://pokutta.com/blog/research/2022/02/20/CGAVI.html} } - Criado, F., Martínez-Rubio, D., and Pokutta, S. (2022). Fast Algorithms for Packing Proportional Fairness and its Dual. Advances in Neural Information Processing Systems, 35, 25754–25766.
Associated with NeurIPS 2022.
DOI: 10.52202/068431-1867
[arXiv]
[poster]
[BibTeX]
@inproceedings{2021_CriadoMartinezrubioPokutta_Packingproportionalfairness, year = {2022}, booktitle = {Advances in Neural Information Processing Systems}, series = {Advances in Neural Information Processing Systems}, volume = {35}, publisher = {Curran Associates}, pages = {25754--25766}, doi = {10.52202/068431-1867}, archiveprefix = {arXiv}, eprint = {2109.03678}, arxiv = {arXiv:2109.03678}, primaryclass = {math.OC}, author = {Criado, Francisco and Martínez-Rubio, David and Pokutta, Sebastian}, title = {Fast Algorithms for Packing Proportional Fairness and its Dual}, poster = {https://pokutta.com/slides/20211105_fairpacking-poster.pdf} } - Carderera, A., Besançon, M., and Pokutta, S. (2021). Simple steps are all you need: Frank-Wolfe and generalized self-concordant functions. Advances in Neural Information Processing Systems, 34, 5390–5401.
Associated with NeurIPS 2021.
[URL]
[arXiv]
[summary]
[slides]
[poster]
[code]
[BibTeX]
@inproceedings{2024_CardereraBesanconPokutta_Scalablefrankwolfe:1, year = {2021}, booktitle = {Advances in Neural Information Processing Systems}, series = {Advances in Neural Information Processing Systems}, volume = {34}, publisher = {Curran Associates}, month = may, pages = {5390--5401}, url = {https://proceedings.neurips.cc/paper_files/paper/2021/file/2b323d6eb28422cef49b266557dd31ad-Paper.pdf}, archiveprefix = {arXiv}, eprint = {2105.13913}, arxiv = {arXiv:2105.13913}, primaryclass = {math.OC}, author = {Carderera, Alejandro and Besançon, Mathieu and Pokutta, Sebastian}, title = {Simple steps are all you need: Frank-Wolfe and generalized self-concordant functions}, code = {https://doi.org/10.5281/zenodo.4836009}, poster = {https://pokutta.com/slides/20211120_poster_NeurIPS21_lSimple_steps_are_all_you_need.pdf}, slides = {https://pokutta.com/slides/20210710_FW-simpleSteps-SelfConcordance.pdf}, summary = {https://pokutta.com/blog/research/2021/10/09/self-concordant-abstract.html} } - Carderera, A., Diakonikolas, J., Lin, C. Y., and Pokutta, S. (2021). Parameter-free Locally Accelerated Conditional Gradients. Proceedings of the 38th International Conference on Machine Learning, 139.
Associated with ICML 2021.
[arXiv]
[slides]
[BibTeX]
@inproceedings{2021_CardereraDiakonikolasLinPokutta_ParameterfreeLocallyAccelerated, year = {2021}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, series = {Proceedings of Machine Learning Research}, volume = {139}, month = feb, archiveprefix = {arXiv}, eprint = {2102.06806}, arxiv = {arXiv:2102.06806}, primaryclass = {math.OC}, author = {Carderera, Alejandro and Diakonikolas, Jelena and Lin, Cheuk Yin and Pokutta, Sebastian}, title = {Parameter-free Locally Accelerated Conditional Gradients}, slides = {https://pokutta.com/slides/20210716_PF_LaCG_Poster.pdf} } - Kerdreux, T., d’Aspremont, A., and Pokutta, S. (2021). Projection-Free Optimization on Uniformly Convex Sets. Proceedings of The 24th International Conference on Artificial Intelligence and Statistics, 130, 19–27.
Associated with AISTATS 2021.
[arXiv]
[slides]
[BibTeX]
@inproceedings{2021_KerdrexDaspremontPokutta_ProjectionFreeOptimization, year = {2021}, booktitle = {Proceedings of The 24th International Conference on Artificial Intelligence and Statistics}, series = {Proceedings of Machine Learning Research}, volume = {130}, month = jan, pages = {19--27}, note = {Oral Presentation + Conference Proceedings}, archiveprefix = {arXiv}, eprint = {2004.11053}, arxiv = {arXiv:2004.11053}, primaryclass = {math.OC}, author = {Kerdreux, Thomas and d'Aspremont, Alexandre and Pokutta, Sebastian}, title = {Projection-Free Optimization on Uniformly Convex Sets}, slides = {https://app.box.com/s/36wj0o8le96rrfdxec774wk7rrdp2vlm} } - Combettes, C., and Pokutta, S. (2020). Boosting Frank-Wolfe by Chasing Gradients. Proceedings of the 37th International Conference on Machine Learning, 119.
Associated with ICML 2020.
[URL]
[arXiv]
[slides]
[code]
[video]
[BibTeX]
@inproceedings{2020_CombettesPokutta_BoostingFrankWolfe, year = {2020}, booktitle = {Proceedings of the 37th International Conference on Machine Learning}, series = {Proceedings of Machine Learning Research}, volume = {119}, month = mar, url = {https://proceedings.mlr.press/v119/combettes20a.html}, archiveprefix = {arXiv}, eprint = {2003.06369}, arxiv = {arXiv:2003.06369}, primaryclass = {math.OC}, author = {Combettes, Cyrille and Pokutta, Sebastian}, title = {Boosting Frank-Wolfe by Chasing Gradients}, code = {https://github.com/cyrillewcombettes/boostfw}, slides = {https://app.box.com/s/wwj247r5d456q0778p9b9y1jm6txuifb}, video = {https://youtube.com/watch?v=BfyV0C5FRbE} } - Diakonikolas, J., Carderera, A., and Pokutta, S. (2020). Locally Accelerated Conditional Gradients. Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics, 108.
Associated with AISTATS 2020.
[URL]
[arXiv]
[slides]
[code]
[BibTeX]
@inproceedings{2019_DiakonikolasCardereraPokutta_BreakingCurse, year = {2020}, booktitle = {Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics}, series = {Proceedings of Machine Learning Research}, volume = {108}, url = {https://proceedings.mlr.press/v108/diakonikolas20a/diakonikolas20a.pdf}, archiveprefix = {arXiv}, eprint = {1906.07867}, arxiv = {arXiv:1906.07867}, primaryclass = {math.OC}, author = {Diakonikolas, Jelena and Carderera, Alejandro and Pokutta, Sebastian}, title = {Locally Accelerated Conditional Gradients}, code = {https://colab.research.google.com/drive/1ejjfCan7xnEhWWJXCIzb03CwQRG9iW_O}, slides = {https://app.box.com/s/gphkhapso7d1vrfnzqykkb3vx0agxh8w} } - Pokutta, S., Singh, M., and Torrico, A. (2020). On the Unreasonable Effectiveness of the Greedy Algorithm: Greedy Adapts to Sharpness. Proceedings of the 37th International Conference on Machine Learning, 119.
Associated with ICML 2020.
[arXiv]
[slides]
[video]
[BibTeX]
@inproceedings{2019_PokuttaSinghTorrico_UnreasonableEffectiveness, year = {2020}, booktitle = {Proceedings of the 37th International Conference on Machine Learning}, series = {Proceedings of Machine Learning Research}, volume = {119}, archiveprefix = {arXiv}, eprint = {2002.04063}, arxiv = {arXiv:2002.04063}, primaryclass = {cs.DS}, author = {Pokutta, Sebastian and Singh, Mohit and Torrico, Alfredo}, title = {On the Unreasonable Effectiveness of the Greedy Algorithm: Greedy Adapts to Sharpness}, slides = {https://app.box.com/s/fmjxhfdpiqubm28upf563ohy8t8fgl3a}, video = {https://youtube.com/watch?v=VB1e0HrDmVo} } - Braun, G., Pokutta, S., Tu, D., and Wright, S. (2019). Blended Conditional Gradients: the unconditioning of conditional gradients. Proceedings of the 36th International Conference on Machine Learning, 97, 735–743.
Associated with ICML 2019.
[URL]
[arXiv]
[summary]
[slides]
[poster]
[code]
[BibTeX]
@inproceedings{2018_BraunPokuttaTuStephen_Blendedconditionalgradients, year = {2019}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, series = {Proceedings of Machine Learning Research}, volume = {97}, pages = {735--743}, url = {https://proceedings.mlr.press/v97/braun19a}, archiveprefix = {arXiv}, eprint = {1805.07311}, arxiv = {arXiv:1805.07311}, primaryclass = {math.OC}, author = {Braun, Gábor and Pokutta, Sebastian and Tu, Dan and Wright, Stephen}, title = {Blended Conditional Gradients: the unconditioning of conditional gradients}, code = {https://github.com/pokutta/bcg}, poster = {https://app.box.com/s/nmmm671jd72i397nysa8emfnzh1hn6hf}, slides = {https://app.box.com/s/xbx3z7ws6dxvl3rzgj4jp6forigycooe}, summary = {https://pokutta.com/blog/research/2019/02/18/bcg-abstract.html} } - Combettes, C., and Pokutta, S. (2019). Blended Matching Pursuit. Advances in Neural Information Processing Systems, 32, 2042–2052.
Associated with NeurIPS 2019.
[URL]
[arXiv]
[slides]
[code]
[BibTeX]
@inproceedings{2019_CombettesPokutta_BlendedMatching, year = {2019}, booktitle = {Advances in Neural Information Processing Systems}, series = {Advances in Neural Information Processing Systems}, volume = {32}, publisher = {Curran Associates}, pages = {2042--2052}, url = {https://papers.nips.cc/paper/8478-blended-matching-pursuit}, archiveprefix = {arXiv}, eprint = {1904.12335}, arxiv = {arXiv:1904.12335}, primaryclass = {math.OC}, author = {Combettes, Cyrille and Pokutta, Sebastian}, title = {Blended Matching Pursuit}, code = {https://colab.research.google.com/drive/17XYIxnCcJjKswba9mAaXFWnNGVZdsaXQ}, slides = {https://app.box.com/s/8lfktq6h3dqp9t2gqydu2tp8h2uxgz7m} } - Diakonikolas, J., Carderera, A., and Pokutta, S. (2019). Breaking the Curse of Dimensionality (Locally) to Accelerate Conditional Gradients. Accepted at Optimization for Machine Learning.
[URL]
[arXiv]
[slides]
[code]
[BibTeX]
@misc{2019_DiakonikolasCardereraPokutta_BreakingCurse:1, howpublished = {Accepted at Optimization for Machine Learning}, year = {2019}, url = {https://opt-ml.org/papers/2019/paper_26.pdf}, archiveprefix = {arXiv}, eprint = {1906.07867}, arxiv = {arXiv:1906.07867}, primaryclass = {math.OC}, author = {Diakonikolas, Jelena and Carderera, Alejandro and Pokutta, Sebastian}, title = {Breaking the Curse of Dimensionality (Locally) to Accelerate Conditional Gradients}, code = {https://colab.research.google.com/drive/1ejjfCan7xnEhWWJXCIzb03CwQRG9iW_O}, slides = {https://app.box.com/s/gphkhapso7d1vrfnzqykkb3vx0agxh8w} } - Kerdreux, T., d’Aspremont, A., and Pokutta, S. (2019). Restarting Frank-Wolfe. Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics, 89, 1275–1283.
Associated with AISTATS 2019.
[URL]
[arXiv]
[slides]
[BibTeX]
@inproceedings{2019_KerdrexDaspremontPokutta_RestartingFrankWolfe:1, year = {2019}, booktitle = {Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics}, series = {Proceedings of Machine Learning Research}, volume = {89}, pages = {1275--1283}, url = {https://proceedings.mlr.press/v89/kerdreux19a/kerdreux19a.pdf}, archiveprefix = {arXiv}, eprint = {1810.02429}, arxiv = {arXiv:1810.02429}, primaryclass = {math.OC}, author = {Kerdreux, Thomas and d'Aspremont, Alexandre and Pokutta, Sebastian}, title = {Restarting Frank-Wolfe}, slides = {https://app.box.com/s/prd32r6xmuef2k4gah23rd0egllz9rv5} } - Pokutta, S., Singh, M., and Torrico, A. (2019). On the Unreasonable Effectiveness of the Greedy Algorithm: Greedy Adapts to Sharpness. Accepted at Optimization for Machine Learning.
[URL]
[arXiv]
[slides]
[video]
[BibTeX]
@misc{2019_PokuttaSinghTorrico_UnreasonableEffectiveness:1, howpublished = {Accepted at Optimization for Machine Learning}, year = {2019}, url = {https://opt-ml.org/papers/2019/paper_16.pdf}, archiveprefix = {arXiv}, eprint = {2002.04063}, arxiv = {arXiv:2002.04063}, primaryclass = {cs.DS}, author = {Pokutta, Sebastian and Singh, Mohit and Torrico, Alfredo}, title = {On the Unreasonable Effectiveness of the Greedy Algorithm: Greedy Adapts to Sharpness}, slides = {https://app.box.com/s/fmjxhfdpiqubm28upf563ohy8t8fgl3a}, video = {https://youtube.com/watch?v=VB1e0HrDmVo} } - Braun, G., Pokutta, S., and Zink, D. (2017). Lazifying Conditional Gradient Algorithms. Proceedings of the 34th International Conference on Machine Learning, 70, 566–575.
Associated with ICML 2017.
[URL]
[arXiv]
[slides]
[poster]
[BibTeX]
@inproceedings{2016_BraunPokuttaZink_Lazifyinggradientalgorithms:1, year = {2017}, booktitle = {Proceedings of the 34th International Conference on Machine Learning}, series = {Proceedings of Machine Learning Research}, volume = {70}, pages = {566--575}, url = {https://proceedings.mlr.press/v70/braun17a}, archiveprefix = {arXiv}, eprint = {1610.05120}, arxiv = {arXiv:1610.05120}, primaryclass = {cs.DS}, author = {Braun, Gábor and Pokutta, Sebastian and Zink, Daniel}, title = {Lazifying Conditional Gradient Algorithms}, poster = {https://app.box.com/s/lysscdg17ytpz7mqr0tu2djffyqvkl6a}, slides = {https://app.box.com/s/zsp0hixjz2ha23u1vuyosijjkjdh8k} } - Lan, G., Pokutta, S., Zhou, Y., and Zink, D. (2017). Conditional Accelerated Lazy Stochastic Gradient Descent. Proceedings of the 34th International Conference on Machine Learning, 70, 1965–1974.
Associated with ICML 2017.
[URL]
[arXiv]
[BibTeX]
@inproceedings{2017_LanPokuttaZhouZink_ConditionalAccelerated, year = {2017}, booktitle = {Proceedings of the 34th International Conference on Machine Learning}, series = {Proceedings of Machine Learning Research}, volume = {70}, pages = {1965--1974}, url = {https://proceedings.mlr.press/v70/lan17a.html}, archiveprefix = {arXiv}, eprint = {1703.05840}, arxiv = {arXiv:1703.05840}, primaryclass = {cs.LG}, author = {Lan, G. and Pokutta, Sebastian and Zhou, Y and Zink, Daniel}, title = {Conditional Accelerated Lazy Stochastic Gradient Descent} } - Roy, A., and Pokutta, S. (2016). Hierarchical Clustering via Spreading Metrics. Advances in Neural Information Processing Systems, 29.
Associated with NeurIPS 2016.
[URL]
[arXiv]
[BibTeX]
@inproceedings{2017_RoyPokutta_HierarchicalClustering:1, year = {2016}, booktitle = {Advances in Neural Information Processing Systems}, series = {Advances in Neural Information Processing Systems}, volume = {29}, publisher = {Curran Associates}, note = {Oral Presentation + Conference Proceedings}, url = {https://papers.nips.cc/paper/by-source-2016-1199}, archiveprefix = {arXiv}, eprint = {1610.09269}, arxiv = {arXiv:1610.09269}, primaryclass = {cs.LG}, author = {Roy, Aurko and Pokutta, Sebastian}, title = {Hierarchical Clustering via Spreading Metrics} } - Braun, G., Pokutta, S., and Xie, Y. (2014). Info-greedy sequential adaptive compressed sensing. 2014 52nd Annual Allerton Conference on Communication, Control, and Computing (Allerton), 858–865.
Associated with Allerton 2014.
DOI: 10.1109/ALLERTON.2014.7028544
[arXiv]
[BibTeX]
@inproceedings{2014_BraunPokuttaYao_Infogreedycompressedsensing:1, year = {2014}, booktitle = {2014 52nd Annual Allerton Conference on Communication, Control, and Computing (Allerton)}, publisher = {Institute of Electrical and Electronics Engineers}, pages = {858--865}, doi = {10.1109/ALLERTON.2014.7028544}, archiveprefix = {arXiv}, eprint = {1407.0731}, arxiv = {arXiv:1407.0731}, primaryclass = {cs.IT}, author = {Braun, Gábor and Pokutta, Sebastian and Xie, Yao}, title = {Info-greedy sequential adaptive compressed sensing} }
Articles, chapters and books
- Wirth, E., Peña, J., and Pokutta, S. (2026). Adaptive Open-Loop Step-Sizes for Accelerated Convergence Rates of the Frank-Wolfe Algorithm. Computational Optimization and Applications.
DOI: 10.1007/s10589-026-00817-w
[arXiv]
[BibTeX]
@article{2025_WirthJavierPokutta_Adaptivestepsizes, year = {2026}, date = {2026-08-11}, journal = {Computational Optimization and Applications}, month = aug, doi = {10.1007/s10589-026-00817-w}, archiveprefix = {arXiv}, eprint = {2505.09886}, arxiv = {arXiv:2505.09886}, primaryclass = {math.OC}, author = {Wirth, Elias and Peña, Javier and Pokutta, Sebastian}, title = {Adaptive Open-Loop Step-Sizes for Accelerated Convergence Rates of the Frank-Wolfe Algorithm} } - Wirth, E., Peña, J., and Pokutta, S. (2026). Fast Convergence of Frank-Wolfe Algorithms on Polytopes. Mathematics of Operations Research, 51(2), 1463–1485.
DOI: 10.1287/moor.2024.0580
[URL]
[arXiv]
[BibTeX]
@article{2024_WirthJavierPokutta_Frankwolfeconvergence, year = {2026}, journal = {Mathematics of Operations Research}, month = may, volume = {51}, number = {2}, pages = {1463--1485}, doi = {10.1287/moor.2024.0580}, url = {https://pubsonline.informs.org/doi/abs/10.1287/moor.2024.0580}, archiveprefix = {arXiv}, eprint = {2406.18789}, arxiv = {arXiv:2406.18789}, primaryclass = {math.OC}, author = {Wirth, Elias and Peña, Javier and Pokutta, Sebastian}, title = {Fast Convergence of Frank-Wolfe Algorithms on Polytopes} } - Kreimeier, T., Pokutta, S., Walther, A., and Woodstock, Z. (2026). On a Frank-Wolfe approach for abs-smooth functions. Optimization Methods & Software, 41(2), 423–449.
DOI: 10.1080/10556788.2023.2296985
[arXiv]
[BibTeX]
@article{2023_KreimeierPokuttaWaltherWoodstock_Frankwolfeabssmooth, year = {2026}, date = {2026-04-29}, journal = {Optimization Methods & Software}, month = apr, volume = {41}, number = {2}, pages = {423--449}, doi = {10.1080/10556788.2023.2296985}, archiveprefix = {arXiv}, eprint = {2303.09881}, arxiv = {arXiv:2303.09881}, primaryclass = {math.OC}, author = {Kreimeier, Timo and Pokutta, Sebastian and Walther, Andrea and Woodstock, Zev}, title = {On a Frank-Wolfe approach for abs-smooth functions} } - Carderera, A., and Pokutta, S. (2026). Second-order Conditional Gradient Sliding. In Data Science and Optimization (Vol. 91, pp. 303–321). Springer Science+Business Media.
DOI: 10.1007/978-3-032-03844-9_12
[arXiv]
[code]
[BibTeX]
@incollection{2020_CarderaPokutta_SecondOrderConditional, year = {2026}, booktitle = {Data Science and Optimization}, series = {Fields Institute Communications}, volume = {91}, publisher = {Springer Science+Business Media}, pages = {303--321}, doi = {10.1007/978-3-032-03844-9_12}, archiveprefix = {arXiv}, eprint = {2002.08907}, arxiv = {arXiv:2002.08907}, primaryclass = {math.OC}, author = {Carderera, Alejandro and Pokutta, Sebastian}, title = {Second-order Conditional Gradient Sliding}, code = {https://github.com/pokutta/Second-order-Conditional-Gradients} } - Kerdreux, T., d’Aspremont, A., and Pokutta, S. (2026). Local and Global Uniform Convexity Conditions. In Data Science and Optimization (Vol. 91, pp. 85–120). Springer Science+Business Media.
[arXiv]
[BibTeX]
@incollection{2021_KerdrexDaspremontPokutta_LocalGlobalUniform, year = {2026}, booktitle = {Data Science and Optimization}, series = {Fields Institute Communications}, volume = {91}, publisher = {Springer Science+Business Media}, pages = {85--120}, archiveprefix = {arXiv}, eprint = {2102.05134}, arxiv = {arXiv:2102.05134}, primaryclass = {math.OC}, author = {Kerdreux, Thomas and d'Aspremont, Alexandre and Pokutta, Sebastian}, title = {Local and Global Uniform Convexity Conditions} } - Braun, G., Halbey, J., Pokutta, S., and Woodstock, Z. (2026). Flexible block-iterative analysis for the Frank-Wolfe algorithm. Journal of Optimization Theory and Applications, 210(37).
DOI: 10.1007/s10957-026-03068-1
[URL]
[arXiv]
[BibTeX]
@article{2024_BraunHalbeyPokuttaWoodstock_Blockiterativeanalysis, year = {2026}, date = {2026-07-31}, journal = {Journal of Optimization Theory and Applications}, volume = {210}, number = {37}, doi = {10.1007/s10957-026-03068-1}, url = {https://link.springer.com/article/10.1007/s10957-026-03068-1}, archiveprefix = {arXiv}, eprint = {2409.06931}, arxiv = {arXiv:2409.06931}, primaryclass = {math.OC}, author = {Braun, Gábor and Halbey, Jannis and Pokutta, Sebastian and Woodstock, Zev}, title = {Flexible block-iterative analysis for the Frank-Wolfe algorithm} } - Besançon, M., Designolle, S., Halbey, J., Hendrych, D., Kuzinowicz, D., Pokutta, S., Troppens, H., Viladrich Herrmannsdoerfer, D., and Wirth, E. (2025). Improved Algorithms and Novel Applications of the FrankWolfe.jl Library. ACM Transactions on Mathematical Software, 51(4), 1–33.
DOI: 10.1145/3765626
[URL]
[arXiv]
[BibTeX]
@article{2025_BesanconEtAl_Frankwolfelibrary, year = {2025}, date = {2025-12-12}, journal = {ACM Transactions on Mathematical Software}, month = dec, volume = {51}, number = {4}, pages = {1--33}, doi = {10.1145/3765626}, url = {https://dl.acm.org/doi/10.1145/3765626}, archiveprefix = {arXiv}, eprint = {2501.14613}, arxiv = {arXiv:2501.14613}, primaryclass = {math.OC}, author = {Besançon, Mathieu and Designolle, Sébastien and Halbey, Jannis and Hendrych, Deborah and Kuzinowicz, Dominik and Pokutta, Sebastian and Troppens, Hannah and Viladrich Herrmannsdoerfer, Daniel and Wirth, Elias}, title = {Improved Algorithms and Novel Applications of the FrankWolfe.jl Library} } - Wirth, E., Pena, J., and Pokutta, S. (2025). Accelerated affine-invariant convergence rates of the Frank–Wolfe algorithm with open-loop step-sizes. Mathematical Programming, 214(1-2), 201–245.
DOI: 10.1007/s10107-024-02180-2
[arXiv]
[BibTeX]
@article{2023_WirthJavierPokutta_Affineinvariantconvergence, year = {2025}, journal = {Mathematical Programming}, month = nov, volume = {214}, number = {1-2}, pages = {201--245}, doi = {10.1007/s10107-024-02180-2}, archiveprefix = {arXiv}, eprint = {2310.04096}, arxiv = {arXiv:2310.04096}, primaryclass = {math.OC}, author = {Wirth, Elias and Pena, Javier and Pokutta, Sebastian}, title = {Accelerated affine-invariant convergence rates of the Frank–Wolfe algorithm with open-loop step-sizes} } - Woodstock, Z., and Pokutta, S. (2025). Splitting the conditional gradient algorithm. SIAM Journal on Optimization, 35(1), 347–368.
DOI: 10.1137/24M1638008
[arXiv]
[BibTeX]
@article{2023_WoodstockPokutta_Conditionalgradientnonconvex, year = {2025}, journal = {SIAM Journal on Optimization}, month = nov, volume = {35}, number = {1}, pages = {347--368}, doi = {10.1137/24M1638008}, archiveprefix = {arXiv}, eprint = {2311.05381}, arxiv = {arXiv:2311.05381}, primaryclass = {math.OC}, author = {Woodstock, Zev and Pokutta, Sebastian}, title = {Splitting the conditional gradient algorithm} } - Carderera, A., Pokutta, S., Schütte, C., and Weiser, M. (2025). An efficient first-order conditional gradient algorithm in data-driven sparse identification of nonlinear dynamics to solve sparse recovery problems under noise. Journal of Computational and Applied Mathematics, 470, 116675.
DOI: 10.1016/j.cam.2025.116675
[URL]
[arXiv]
[BibTeX]
@article{2021_CardereraPokuttaSchutteWeiser_CINDy, year = {2025}, date = {2025-04-01}, journal = {Journal of Computational and Applied Mathematics}, month = apr, volume = {470}, pages = {116675}, doi = {10.1016/j.cam.2025.116675}, url = {https://www.sciencedirect.com/science/article/abs/pii/S037704272500189X}, archiveprefix = {arXiv}, eprint = {2101.02630}, arxiv = {arXiv:2101.02630}, primaryclass = {math.DS}, author = {Carderera, Alejandro and Pokutta, Sebastian and Schütte, 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} } - Braun, G., Carderera, A., Combettes, C., Hassani, H., Karbasi, A., Mokhtari, A., and Pokutta, S. (2025). Conditional Gradient Methods: From Core Principles to AI Applications. Society for Industrial and Applied Mathematics.
DOI: 10.1137/1.9781611978568
[arXiv]
[BibTeX]
@book{2022_BraunEtAl_Conditionalgradient, year = {2025}, series = {MOS-SIAM Series on Optimization}, publisher = {Society for Industrial and Applied Mathematics}, doi = {10.1137/1.9781611978568}, archiveprefix = {arXiv}, eprint = {2211.14103}, arxiv = {arXiv:2211.14103}, primaryclass = {math.OC}, author = {Braun, Gábor and Carderera, Alejandro and Combettes, Cyrille and Hassani, Hamed and Karbasi, Amin and Mokhtari, Aryan and Pokutta, Sebastian}, title = {Conditional Gradient Methods: From Core Principles to AI Applications} } - Carderera, A., Besançon, M., and Pokutta, S. (2024). Scalable Frank-Wolfe on generalized self-concordant functions via simple steps. SIAM Journal on Optimization, 34(3), 2231–2258.
DOI: 10.1137/23M1616789
[URL]
[arXiv]
[summary]
[slides]
[code]
[BibTeX]
@article{2024_CardereraBesanconPokutta_Scalablefrankwolfe, year = {2024}, date = {2024-07-02}, journal = {SIAM Journal on Optimization}, month = sep, volume = {34}, number = {3}, pages = {2231--2258}, doi = {10.1137/23M1616789}, url = {https://epubs.siam.org/doi/full/10.1137/23M1616789?journalCode=sjope8}, archiveprefix = {arXiv}, eprint = {2105.13913}, arxiv = {arXiv:2105.13913}, primaryclass = {math.OC}, author = {Carderera, Alejandro and Besançon, Mathieu and Pokutta, Sebastian}, title = {Scalable Frank-Wolfe on generalized self-concordant functions via simple steps}, code = {https://doi.org/10.5281/zenodo.4836009}, slides = {https://pokutta.com/slides/20210710_FW-simpleSteps-SelfConcordance.pdf}, summary = {https://pokutta.com/blog/research/2021/10/09/self-concordant-abstract.html} } - Braun, G., Guzmán, C., and Pokutta, S. (2024). Corrections to “Lower Bounds on the Oracle Complexity of Nonsmooth Convex Optimization via Information Theory.” IEEE Transactions on Information Theory, 70(7), 5408–5409.
DOI: 10.1109/TIT.2024.3357200
[BibTeX]
@article{2024_BraunGuzmanPokutta_Oraclecomplexity, year = {2024}, date = {2024-06-21}, journal = {IEEE Transactions on Information Theory}, month = jul, volume = {70}, number = {7}, pages = {5408--5409}, doi = {10.1109/TIT.2024.3357200}, author = {Braun, Gábor and Guzmán, Cristóbal and Pokutta, Sebastian}, title = {Corrections to “Lower Bounds on the Oracle Complexity of Nonsmooth Convex Optimization via Information Theory”} } - Besançon, M., Dias Garcia, J., Legat, B., and Sharma, A. (2024). Flexible Differentiable Optimization via Model Transformations. INFORMS Journal on Computing, 36(2), 456–478.
DOI: 10.1287/ijoc.2022.0283
[URL]
[arXiv]
[BibTeX]
@article{2022_BesanconJoaquimBenotAkshay_Differentiableoptimization, year = {2024}, journal = {INFORMS Journal on Computing}, month = mar, volume = {36}, number = {2}, pages = {456--478}, doi = {10.1287/ijoc.2022.0283}, url = {https://pubsonline.informs.org/doi/epdf/10.1287/ijoc.2022.0283}, archiveprefix = {arXiv}, eprint = {2206.06135}, arxiv = {arXiv:2206.06135}, primaryclass = {cs.LG}, author = {Besançon, Mathieu and Dias Garcia, Joaquim and Legat, Benoît and Sharma, Akshay}, title = {Flexible Differentiable Optimization via Model Transformations} } - Pokutta, S. (2024). The Frank-Wolfe Algorithm: A Short Introduction. Jahresbericht Der Deutschen Mathematiker-Vereinigung, 126(1), 3–35.
DOI: 10.1365/s13291-023-00275-x
[URL]
[arXiv]
[BibTeX]
@article{2023_Pokutta_Frankwolfealgorithm, year = {2024}, journal = {Jahresbericht der Deutschen Mathematiker-Vereinigung}, month = mar, volume = {126}, number = {1}, pages = {3--35}, doi = {10.1365/s13291-023-00275-x}, url = {https://link.springer.com/article/10.1365/s13291-023-00275-x/fulltext.html}, archiveprefix = {arXiv}, eprint = {2311.05313}, arxiv = {arXiv:2311.05313}, primaryclass = {math.OC}, author = {Pokutta, Sebastian}, title = {The Frank-Wolfe Algorithm: A Short Introduction} } - Hunkenschröder, C., Pokutta, S., and Weismantel, R. (2023). Minimizing a low-dimensional convex function over a high-dimensional cube. SIAM Journal on Optimization, 33(2), 538–552.
DOI: 10.1137/22M1489988
[arXiv]
[BibTeX]
@article{2022_HunkenschrderPokuttaWeismantel_Convexoptimization, year = {2023}, date = {2023-05-26}, journal = {SIAM Journal on Optimization}, month = jun, volume = {33}, number = {2}, pages = {538--552}, doi = {10.1137/22M1489988}, archiveprefix = {arXiv}, eprint = {2204.05266}, arxiv = {arXiv:2204.05266}, primaryclass = {math.OC}, author = {Hunkenschröder, Christoph and Pokutta, Sebastian and Weismantel, Robert}, title = {Minimizing a low-dimensional convex function over a high-dimensional cube} } - Combettes, C., and Pokutta, S. (2023). Revisiting the approximate Carathéodory problem via the Frank-Wolfe algorithm. Mathematical Programming, 197(1), 191–214.
DOI: 10.1007/s10107-021-01735-x
[URL]
[arXiv]
[slides]
[code]
[video]
[BibTeX]
@article{2023_CombettesPokutta_RevisitingApproximateCaratheodory, year = {2023}, journal = {Mathematical Programming}, volume = {197}, number = {1}, pages = {191--214}, doi = {10.1007/s10107-021-01735-x}, url = {https://rdcu.be/cCnPL}, archiveprefix = {arXiv}, eprint = {1911.04415}, arxiv = {arXiv:1911.04415}, primaryclass = {math.OC}, author = {Combettes, Cyrille and Pokutta, Sebastian}, title = {Revisiting the approximate Carathéodory problem via the Frank-Wolfe algorithm}, code = {https://colab.research.google.com/drive/1GLGRTc2jFYy9CqqoVnZgIFVQgAC0c3aZ}, slides = {https://app.box.com/s/f0zuvr45qa6etidd06i1wart58o7bc8t}, video = {https://youtube.com/watch?v=VB1e0HrDmVo} } - Wilken, S. E., Besançon, M., Kratochvíl, M., Foko Kuate, C. A., Trefois, C., Gu, W., and Ebenhöh, O. (2022). Interrogating the effect of enzyme kinetics on metabolism using differentiable constraint-based models. Metabolic Engineering, 74, 72–82.
DOI: 10.1016/j.ymben.2022.09.002
[BibTeX]
@article{2022_ElmoEtAl_Enzymekineticsmetabolism, year = {2022}, journal = {Metabolic Engineering}, month = nov, volume = {74}, pages = {72--82}, doi = {10.1016/j.ymben.2022.09.002}, author = {Wilken, St. Elmo and Besançon, Mathieu and Kratochvíl, Miroslav and Foko Kuate, Chilperic Armel and Trefois, Christophe and Gu, Wei and Ebenhöh, Oliver}, title = {Interrogating the effect of enzyme kinetics on metabolism using differentiable constraint-based models} } - Besançon, M., Carderera, A., and Pokutta, S. (2022). FrankWolfe.jl: A High-Performance and Flexible Toolbox for Frank–Wolfe Algorithms and Conditional Gradients. INFORMS Journal on Computing, 34(5), 2611–2620.
DOI: 10.1287/ijoc.2022.1191
[URL]
[arXiv]
[summary]
[slides]
[code]
[BibTeX]
@article{2021_BesanconCardereraPokutta_Frankwolfetoolbox, year = {2022}, journal = {INFORMS Journal on Computing}, month = sep, volume = {34}, number = {5}, pages = {2611--2620}, doi = {10.1287/ijoc.2022.1191}, url = {https://pubsonline.informs.org/doi/abs/10.1287/ijoc.2022.1191}, archiveprefix = {arXiv}, eprint = {2104.06675}, arxiv = {arXiv:2104.06675}, primaryclass = {math.OC}, author = {Besançon, Mathieu and Carderera, Alejandro and Pokutta, Sebastian}, title = {FrankWolfe.jl: A High-Performance and Flexible Toolbox for Frank–Wolfe Algorithms and Conditional Gradients}, code = {https://github.com/ZIB-IOL/FrankWolfe.jl}, slides = {https://pokutta.com/slides/20210710_FW-simpleSteps-SelfConcordance.pdf}, summary = {https://pokutta.com/blog/research/2021/04/20/FrankWolfejl.html} } - Kerdreux, T., d’Aspremont, A., and Pokutta, S. (2022). Restarting Frank–Wolfe: Faster Rates under Hölderian Error Bounds. Journal of Optimization Theory and Applications, 192(3), 799–829.
DOI: 10.1007/s10957-021-01989-7
[URL]
[arXiv]
[slides]
[BibTeX]
@article{2019_KerdrexDaspremontPokutta_RestartingFrankWolfe, year = {2022}, journal = {Journal of Optimization Theory and Applications}, volume = {192}, number = {3}, pages = {799--829}, doi = {10.1007/s10957-021-01989-7}, url = {https://dx.doi.org/10.1007/s10957-021-01989-7}, archiveprefix = {arXiv}, eprint = {1810.02429}, arxiv = {arXiv:1810.02429}, primaryclass = {math.OC}, author = {Kerdreux, Thomas and d'Aspremont, Alexandre and Pokutta, Sebastian}, title = {Restarting Frank–Wolfe: Faster Rates under Hölderian Error Bounds}, slides = {https://app.box.com/s/prd32r6xmuef2k4gah23rd0egllz9rv5} } - Combettes, C., and Pokutta, S. (2021). Complexity of linear minimization and projection on some sets. Operations Research Letters, 49(4), 565–571.
DOI: 10.1016/j.orl.2021.06.005
[arXiv]
[code]
[BibTeX]
@article{2021_CombettesPokutta_ComplexityLinearMinimization, year = {2021}, journal = {Operations Research Letters}, month = jul, volume = {49}, number = {4}, pages = {565--571}, doi = {10.1016/j.orl.2021.06.005}, archiveprefix = {arXiv}, eprint = {2101.10040}, arxiv = {arXiv:2101.10040}, primaryclass = {math.OC}, author = {Combettes, Cyrille and Pokutta, Sebastian}, title = {Complexity of linear minimization and projection on some sets}, code = {https://github.com/cyrillewcombettes/complexity} } - Kerdreux, T., Roux, C., d’Aspremont, A., and Pokutta, S. (2021). Linear Bandits on Uniformly Convex Sets. Journal of Machine Learning Research, 22(284), 1–23.
[URL]
[arXiv]
[summary]
[BibTeX]
@article{2021_KerdreuxRouxDaspremontPokutta_Linearbandits, year = {2021}, journal = {Journal of Machine Learning Research}, month = mar, volume = {22}, number = {284}, pages = {1--23}, url = {https://jmlr.org/papers/v22/21-0277.html}, archiveprefix = {arXiv}, eprint = {2103.05907}, arxiv = {arXiv:2103.05907}, primaryclass = {cs.LG}, author = {Kerdreux, Thomas and Roux, Christophe and d'Aspremont, Alexandre and Pokutta, Sebastian}, title = {Linear Bandits on Uniformly Convex Sets}, summary = {https://www.pokutta.com/blog/research/2021/04/03/linearBandits.html} } - Braun, G., Pokutta, S., and Zink, D. (2019). Lazifying Conditional Gradient Algorithms. Journal of Machine Learning Research, 20(71), 1–42.
[URL]
[arXiv]
[BibTeX]
@article{2016_BraunPokuttaZink_Lazifyinggradientalgorithms, year = {2019}, journal = {Journal of Machine Learning Research}, volume = {20}, number = {71}, pages = {1--42}, url = {https://jmlr.org/papers/v20/18-114.html}, archiveprefix = {arXiv}, eprint = {1610.05120}, arxiv = {arXiv:1610.05120}, primaryclass = {cs.DS}, author = {Braun, Gábor and Pokutta, Sebastian and Zink, Daniel}, title = {Lazifying Conditional Gradient Algorithms} } - Braun, G., Guzmán, C., and Pokutta, S. (2017). Lower Bounds on the Oracle Complexity of Nonsmooth Convex Optimization via Information Theory. IEEE Transactions on Information Theory, 63(7), 4709–4724.
DOI: 10.1109/TIT.2017.2701343
[URL]
[arXiv]
[BibTeX]
@article{2014_BraunGuzmanPokutta_Lowerboundsoraclecomplexity, year = {2017}, journal = {IEEE Transactions on Information Theory}, month = jul, volume = {63}, number = {7}, pages = {4709--4724}, doi = {10.1109/TIT.2017.2701343}, url = {https://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=7919238}, archiveprefix = {arXiv}, eprint = {1407.5144}, arxiv = {arXiv:1407.5144}, primaryclass = {math.OC}, author = {Braun, Gábor and Guzmán, Cristóbal and Pokutta, Sebastian}, title = {Lower Bounds on the Oracle Complexity of Nonsmooth Convex Optimization via Information Theory} } - Braun, G., Jain, R., Lee, T., and Pokutta, S. (2017). Information-theoretic approximations of the nonnegative rank. Computational Complexity, 26(1), 147–197.
DOI: 10.1007/s00037-016-0125-z
[URL]
[BibTeX]
@article{2017_BraunRahulTroyPokutta_Informationtheoreticnonnegativerank, year = {2017}, journal = {Computational Complexity}, volume = {26}, number = {1}, pages = {147--197}, doi = {10.1007/s00037-016-0125-z}, url = {https://eccc.weizmann.ac.il/report/2013/158}, author = {Braun, Gábor and Jain, Rahul and Lee, Troy and Pokutta, Sebastian}, title = {Information-theoretic approximations of the nonnegative rank} } - Roy, A., and Pokutta, S. (2017). Hierarchical Clustering via Spreading Metrics. Journal of Machine Learning Research, 18, 1–35.
[URL]
[arXiv]
[BibTeX]
@article{2017_RoyPokutta_HierarchicalClustering, year = {2017}, journal = {Journal of Machine Learning Research}, volume = {18}, pages = {1--35}, url = {https://jmlr.org/papers/v18/17-081.html}, archiveprefix = {arXiv}, eprint = {1610.09269}, arxiv = {arXiv:1610.09269}, primaryclass = {cs.LG}, author = {Roy, Aurko and Pokutta, Sebastian}, title = {Hierarchical Clustering via Spreading Metrics} } - Braun, G., Pokutta, S., and Xie, Y. (2015). Info-Greedy Sequential Adaptive Compressed Sensing. IEEE Journal of Selected Topics in Signal Processing, 9(4), 601–611.
DOI: 10.1109/JSTSP.2015.2400428
[arXiv]
[BibTeX]
@article{2014_BraunPokuttaYao_Infogreedycompressedsensing, year = {2015}, journal = {IEEE Journal of Selected Topics in Signal Processing}, month = jun, volume = {9}, number = {4}, pages = {601--611}, doi = {10.1109/JSTSP.2015.2400428}, archiveprefix = {arXiv}, eprint = {1407.0731}, arxiv = {arXiv:1407.0731}, primaryclass = {cs.IT}, author = {Braun, Gábor and Pokutta, Sebastian and Xie, Yao}, title = {Info-Greedy Sequential Adaptive Compressed Sensing} }
Projects
Completed Projects
Motivated by nonsmooth problems in machine learning, this project addresses minimizing abs-smooth functions subject to closed convex constraints. New theory and algorithms are developed using linear minimization oracles to enforce constraints and abs-linearization methods to handle nonsmoothness.
This project studies algorithms that promote sparsity, developing PageRank optimization algorithms that scale with solution sparsity and investigating Riemannian optimization using manifold geometry. It also develops algorithms for efficient fair resource allocation based on established fairness axioms.









