What we are interested in
Deep Learning (DL) has proven to be a powerful tool to address complex problems across a wide range of fields. At iol.LEARN, we mainly focus on the intersection of sustainability and AI from two perspectives: a) enhancing the efficiency of DL methods, and b) employing DL techniques to tackle sustainability-related challenges and other public interest issues. For the former, addressing the challenges posed by the high computational, financial, and environmental costs of training and operating large models such as LLMs, we develop techniques to enhance the computational and memory efficiency through methods like pruning, quantization and decentralized training. For the latter, our group develops methods to tackle sustainability-related challenges, exemplified by our AI4Forest project, which involves learning high-resolution global tree height maps from satellite imagery to monitor deforestation and forest disturbances. We are further interested in taking advantage of DL for scientific discovery (AI4Science) and improving DL training towards robustness and interpretability through theoretically grounded techniques rooted in continuous and discrete optimization.
đ§âđ Members

asadull (at) zib.de

birsan (at) zib.de

chitranshi (at) zib.de

graczyk (at) zib.de

kuzinowicz (at) zib.de

meise (at) zib.de

mundinger (at) zib.de

pelleriti (at) zib.de

prause (at) zib.de

roux (at) zib.de

sadiku (at) zib.de

turan (at) zib.de

urbano (at) zib.de

wagner (at) zib.de
Former members
- Kartikey Sharma (2020â2024)
- Megi Andoni (2023â2024)
- Muhammad Nawfal Meeran (2024)
- Surya Iyer (2022â2023)
- Adnan Mahmud (2023)
- Maria Bulychev (2021â2023)
- Giacomo Salvati (2021â2023)
- Elias Schecke (2021â2022)
- Stephan WĂ€ldchen (2020â2022)
- Sergi Andreu (2020â2021)
- Minh Pham (2020â2021)
- Max-Georg Schorr (2020â2021)
- Margot Cosson (2021)
- Huy Do Anh
Visitors
- Kera Hiroshi (October 2024âSeptember 2025)
- Krunal Patel (February 2024)
đŹ Talks and posters
Conference and workshop talks
- Nov 2024
- A Combinatorial Comparison of Boosting and Repeated Boosting by Ingo Meise
Annual Meeting, SPP Theoretical Foundations of Deep Learning, Tutzing - Oct 2024
- Extending the Continuum of Six-Colorings by Christoph Spiegel
41st Kolloquium ĂŒber Kombinatorik (KolKom), Heidelberg [PDF] - Jul 2024
- Extending the Continuum of Six-Colorings by Christoph Spiegel
13th Discrete Mathematics Days (DMD), AlcalĂĄ de Henares [PDF] - Sep 2022
- Wavelet-based Low Frequency Adversarial Attacks by Shpresim Sadiku
3rd BMS-BGSMath Junior Meeting, Barcelona [PDF] - Mar 2021
- Neural Network Approximation Theory by Shpresim Sadiku
BMS Conference [PDF] - Dec 2019
- Tensor-based Algorithms for Image Classification by Patrick GelĂ
Mathematics of Deep Learning
Research seminar talks
- Dec 2024
- Coloring the Plane with Neural Networks by Christoph Spiegel
Research Seminar Combinatorics, Berlin [PDF] - May 2024
- Learning Operators Via Hypernetworks by Konrad Mundinger
Research Seminar Numerical Analysis of Stochastic and Deterministic Partial Differential Equations, Berlin - Aug 2023
- Minimally Distorted Explainable Adversarial Attacks by Shpresim Sadiku
AIP Seminar, RIKEN Center for Advanced Intelligence Project, Tokyo - Feb 2023
- What Is Backpropagation? by Shpresim Sadiku
"What is ...?" seminar [PDF]
Poster presentations
- Jul 2024
- Estimating Canopy Height at Scale by Max Zimmer
41st International Conference on Machine Learning (ICML), Vienna - Jul 2024
- Unified Taxonomy in AI Safety: Watermarks, Adversarial Defenses, and Transferable Attacks by Berkant Turan
Workshop on Theoretical Foundations of Foundation Models (TF2M) @ ICML 2024, Vienna - May 2024
- Sparse Model Soups a Recipe for Improved Pruning Via Model Averaging by Max Zimmer
12th International Conference on Learning Representations (ICLR), Vienna - May 2024
- Interpretability Guarantees with Merlin-Arthur Classifiers by Berkant Turan
27th AISTATS Conference, ValĂšncia - May 2024
- Neural Parameter Regression for Explicit Representations of PDE Solution Operators by Konrad Mundinger
Workshop on AI4DifferentialEquations in Science @ ICLR 2024, Vienna - Mar 2024
- Group-wise Sparse and Explainable Adversarial Attacks by Shpresim Sadiku
Deep Learning: Theory, Applications, and Implications, RIKEN Center for Advanced Intelligence Project, Tokyo - Jul 2023
- Extending Merlin-Arthur Classifiers for Improved Interpretability by Berkant Turan
The 1st World Conference on eXplainable Artificial Intelligence - May 2023
- How I Learned to Stop Worrying and Love Retraining by Max Zimmer
11th International Conference on Learning Representations (ICLR), Kigali - Mar 2023
- Wavelet-based Low Frequency Adversarial Attacks by Shpresim Sadiku
Workshop on Optimization and Machine Learning, Waischenfeld - Mar 2023
- How I Learned to Stop Worrying and Love Retraining by Max Zimmer
Workshop on Optimization and Machine Learning, Waischenfeld - Dec 2021
- Learning to Schedule Heuristics in Branch-and-Bound by Antonia Chmiela
conference on neural information processing systems (NeurIPS)
đ Publications and preprints
Preprints
- Fayad, I., Zimmer, M., Schwartz, M., Ciais, P., Gieseke, F., Belouze, G., Brood, S., De Truchis, A., and dâAspremont, A. (2025). DUNIA: Pixel-Sized Embeddings Via Cross-Modal Alignment for Earth Observation Applications.
[arXiv]
[BibTeX]
@misc{2025_IbrahimEtAl_DuniaCrossmodalalignment, archiveprefix = {arXiv}, eprint = {2502.17066}, primaryclass = {cs.CV}, year = {2025}, author = {Fayad, Ibrahim and Zimmer, Max and Schwartz, Martin and Ciais, Philippe and Gieseke, Fabian and Belouze, Gabriel and Brood, Sarah and De Truchis, Aurelien and d'Aspremont, Alexandre}, title = {DUNIA: Pixel-Sized Embeddings Via Cross-Modal Alignment for Earth Observation Applications}, date = {2025-02-24} }
- Pauls, J., Zimmer, M., Turan, B., Saatchi, S., Ciais, P., Pokutta, S., and Gieseke, F. (2025). Capturing Temporal Dynamics in Large-Scale Canopy Tree Height Estimation.
[arXiv]
[BibTeX]
@misc{2025_JanEtAl_Temporalcanopyheight, archiveprefix = {arXiv}, eprint = {2501.19328}, primaryclass = {cs.LG}, year = {2025}, author = {Pauls, Jan and Zimmer, Max and Turan, Berkant and Saatchi, Sassan and Ciais, Philippe and Pokutta, Sebastian and Gieseke, Fabian}, title = {Capturing Temporal Dynamics in Large-Scale Canopy Tree Height Estimation}, date = {2025-01-31} }
- Mundinger, K., Zimmer, M., Kiem, A., Spiegel, C., and Pokutta, S. (2025). Neural Discovery in Mathematics: Do Machines Dream of Colored Planes?
[arXiv]
[BibTeX]
@misc{2025_MundingerZimmerKiemSpiegelPokutta_NeuralDiscovery, archiveprefix = {arXiv}, eprint = {2501.18527}, primaryclass = {cs.LG}, year = {2025}, author = {Mundinger, Konrad and Zimmer, Max and Kiem, Aldo and Spiegel, Christoph and Pokutta, Sebastian}, title = {Neural Discovery in Mathematics: Do Machines Dream of Colored Planes?}, date = {2025-01-30} }
- Pelleriti, N., Zimmer, M., Wirth, E., and Pokutta, S. (2025). Approximating Latent Manifolds in Neural Networks Via Vanishing Ideals.
[arXiv]
[BibTeX]
@misc{2025_PelleritiZimmerWirthPokutta_Latentmanifolds, archiveprefix = {arXiv}, eprint = {2502.15051}, primaryclass = {cs.LG}, year = {2025}, author = {Pelleriti, Nico and Zimmer, Max and Wirth, Elias and Pokutta, Sebastian}, title = {Approximating Latent Manifolds in Neural Networks Via Vanishing Ideals}, date = {2025-02-20} }
- GöĂ, A., Martin, A., Pokutta, S., and Sharma, K. (2023). Norm-induced Cuts: Optimization with Lipschitzian Black-box Functions.
[URL]
[arXiv]
[BibTeX]
@misc{2024_AdrianMartinPokuttaSharma_Norminducedcuts, url = {https://opus4.kobv.de/opus4-trr154/files/518/nic_preprint.pdf}, archiveprefix = {arXiv}, eprint = {2403.11546}, primaryclass = {math.OC}, year = {2024}, author = {GöĂ, Adrian and Martin, Alexander and Pokutta, Sebastian and Sharma, Kartikey}, title = {Norm-induced Cuts: Optimization with Lipschitzian Black-box Functions}, date = {2023-05-16} }
- Goerigk, M., Hartisch, M., Merten, S., and Sharma, K. (2024). Feature-Based Interpretable Optimization.
[arXiv]
[BibTeX]
@misc{2024_GoerigkHartischMertenSharma_Featureinterpretableoptimization, archiveprefix = {arXiv}, eprint = {2409.01869}, primaryclass = {math.OC}, year = {2024}, author = {Goerigk, Marc and Hartisch, Michael and Merten, Sebastian and Sharma, Kartikey}, title = {Feature-Based Interpretable Optimization}, date = {2024-09-03} }
- GĆuch, G., Turan, B., Nagarajan, S. G., and Pokutta, S. (2024). The Good, the Bad and the Ugly: Watermarks, Transferable Attacks and Adversarial Defenses.
[arXiv]
[BibTeX]
@misc{2024_GrzegorzTuranNagarajanPokutta_Watermarksadversarialdefenses, archiveprefix = {arXiv}, eprint = {2410.08864}, primaryclass = {cs.LG}, year = {2024}, author = {GĆuch, Grzegorz and Turan, Berkant and Nagarajan, Sai Ganesh and Pokutta, Sebastian}, title = {The Good, the Bad and the Ugly: Watermarks, Transferable Attacks and Adversarial Defenses}, date = {2024-10-11} }
- Mundinger, K., Zimmer, M., and Pokutta, S. (2024). Neural Parameter Regression for Explicit Representations of PDE Solution Operators.
[arXiv]
[BibTeX]
@misc{2024_MundingerZimmerPokutta_Neuralparameterregression, archiveprefix = {arXiv}, eprint = {2403.12764}, primaryclass = {cs.LG}, year = {2024}, author = {Mundinger, Konrad and Zimmer, Max and Pokutta, Sebastian}, title = {Neural Parameter Regression for Explicit Representations of PDE Solution Operators}, date = {2024-03-19} }
- Haase, J., and Pokutta, S. (2024). Human-AI Co-Creativity: Exploring Synergies Across Levels of Creative Collaboration.
[arXiv]
[BibTeX]
@misc{2024_Pokutta_HumanAICOcreative, archiveprefix = {arXiv}, eprint = {2411.12527}, primaryclass = {cs.HC}, year = {2024}, author = {Haase, Jennifer and Pokutta, Sebastian}, title = {Human-AI Co-Creativity: Exploring Synergies Across Levels of Creative Collaboration}, date = {2024-11-19} }
- Zimmer, M., Andoni, M., Spiegel, C., and Pokutta, S. (2023). PERP: Rethinking the Prune-Retrain Paradigm in the Era of LLMs.
[arXiv]
[code]
[BibTeX]
@misc{2023_ZimmerAndoniSpiegelPokutta_PerpPruneRetrain, archiveprefix = {arXiv}, eprint = {2312.15230}, primaryclass = {cs.LG}, year = {2023}, author = {Zimmer, Max and Andoni, Megi and Spiegel, Christoph and Pokutta, Sebastian}, title = {PERP: Rethinking the Prune-Retrain Paradigm in the Era of LLMs}, code = {https://github.com/ZIB-IOL/PERP}, date = {2023-12-23} }
- Zimmer, M., Spiegel, C., and Pokutta, S. (2022). Compression-aware Training of Neural Networks Using Frank-Wolfe.
[arXiv]
[BibTeX]
@misc{2022_ZimmerSpiegelPokutta_Compressionawaretraining, archiveprefix = {arXiv}, eprint = {2205.11921}, primaryclass = {cs.LG}, year = {2022}, author = {Zimmer, Max and Spiegel, Christoph and Pokutta, Sebastian}, title = {Compression-aware Training of Neural Networks Using Frank-Wolfe}, date = {2022-05-24} }
- Carderera, A., Pokutta, S., SchĂŒtte, C., and Weiser, M. (2021). CINDy: Conditional Gradient-based Identification of Non-linear Dynamics â Noise-robust Recovery.
[arXiv]
[BibTeX]
@misc{2021_CardereraPokuttaSchutteWeiser_CINDy, archiveprefix = {arXiv}, eprint = {2101.02630}, primaryclass = {math.DS}, year = {2021}, author = {Carderera, Alejandro and Pokutta, Sebastian and SchĂŒtte, Christof and Weiser, Martin}, title = {CINDy: Conditional Gradient-based Identification of Non-linear Dynamics -- Noise-robust Recovery}, date = {2021-01-07} }
- 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}, primaryclass = {math.OC}, year = {2020}, 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} }
- Pokutta, S., Spiegel, C., and Zimmer, M. (2020). Deep Neural Network Training with Frank-Wolfe.
[arXiv]
[summary]
[code]
[BibTeX]
@misc{2020_PokuttaSpiegelZimmer_Frankwolfeneuralnetworks, archiveprefix = {arXiv}, eprint = {2010.07243}, primaryclass = {cs.LG}, year = {2020}, author = {Pokutta, Sebastian and Spiegel, Christoph and Zimmer, Max}, title = {Deep Neural Network Training with Frank-Wolfe}, code = {https://github.com/ZIB-IOL/StochasticFrankWolfe}, summary = {https://pokutta.com/blog/research/2020/11/11/NNFW.html}, date = {2020-10-14} }
Conference proceedings
- Sadiku, S., Wagner, M., and Pokutta, S. (2023, November 29). Group-wise Sparse and Explainable Adversarial Attacks. Proceedings of the International Conference on Learning Representations.
[arXiv]
[BibTeX]
@inproceedings{2023_SadikuWagnerPokutta_Groupwisesparseattacks, year = {2025}, booktitle = {Proceedings of the International Conference on Learning Representations}, archiveprefix = {arXiv}, eprint = {2311.17434}, primaryclass = {cs.CV}, author = {Sadiku, Shpresim and Wagner, Moritz and Pokutta, Sebastian}, title = {Group-wise Sparse and Explainable Adversarial Attacks}, date = {2023-11-29} }
- Hollender, A., Maystre, G., and Nagarajan, S. G. (2024, September 11). The Complexity of Two-Team Polymatrix Games with Independent Adversaries. Proceedings of the International Conference on Learning Representations.
[arXiv]
[BibTeX]
@inproceedings{2024_HollenderMaystreNagarajan_Twoteampolymatrix, year = {2025}, booktitle = {Proceedings of the International Conference on Learning Representations}, archiveprefix = {arXiv}, eprint = {2409.07398}, primaryclass = {cs.GT}, author = {Hollender, Alexandros and Maystre, Gilbert and Nagarajan, Sai Ganesh}, title = {The Complexity of Two-Team Polymatrix Games with Independent Adversaries}, date = {2024-09-11} }
- Roux, C., Zimmer, M., and Pokutta, S. (2024, February 19). On the Byzantine-resilience of Distillation-based Federated Learning. Proceedings of the International Conference on Learning Representations.
[arXiv]
[BibTeX]
@inproceedings{2024_RouxZimmerPokutta_Byzantineresilience, year = {2025}, booktitle = {Proceedings of the International Conference on Learning Representations}, archiveprefix = {arXiv}, eprint = {2402.12265}, primaryclass = {cs.LG}, author = {Roux, Christophe and Zimmer, Max and Pokutta, Sebastian}, title = {On the Byzantine-resilience of Distillation-based Federated Learning}, date = {2024-02-19} }
- Sadiku, S., Wagner, M., Nagarajan, S. G., and Pokutta, S. (2024, October 21). S-CFE: Simple Counterfactual Explanations. Proceedings of the International Conference on Artificial Intelligence and Statistics.
[arXiv]
[BibTeX]
@inproceedings{2024_SadikuEtAl_Counterfactualexplanations, year = {2025}, booktitle = {Proceedings of the International Conference on Artificial Intelligence and Statistics}, archiveprefix = {arXiv}, eprint = {2410.15723}, primaryclass = {cs.LG}, author = {Sadiku, Shpresim and Wagner, Moritz and Nagarajan, Sai Ganesh and Pokutta, Sebastian}, title = {S-CFE: Simple Counterfactual Explanations}, date = {2024-10-21} }
- WĂ€ldchen, S., Sharma, K., Turan, B., Zimmer, M., and Pokutta, S. (2022, June 1). Interpretability Guarantees with Merlin-Arthur Classifiers. Proceedings of the International Conference on Artificial Intelligence and Statistics.
[arXiv]
[BibTeX]
@inproceedings{2022_WaeldchenEtAl_Interpretabilityguarantees, year = {2024}, booktitle = {Proceedings of the International Conference on Artificial Intelligence and Statistics}, archiveprefix = {arXiv}, eprint = {2206.00759}, primaryclass = {cs.LG}, author = {WĂ€ldchen, Stephan and Sharma, Kartikey and Turan, Berkant and Zimmer, Max and Pokutta, Sebastian}, title = {Interpretability Guarantees with Merlin-Arthur Classifiers}, date = {2022-06-01} }
- Zimmer, M., Spiegel, C., and Pokutta, S. (2023, June 29). Sparse Model Soups: A Recipe for Improved Pruning Via Model Averaging. Proceedings of the International Conference on Learning Representations.
[URL]
[arXiv]
[BibTeX]
@inproceedings{2023_ZimmerSpiegelPokutta_Sparsemodelsoups, year = {2024}, booktitle = {Proceedings of the International Conference on Learning Representations}, url = {https://iclr.cc/virtual/2024/poster/17433}, archiveprefix = {arXiv}, eprint = {2306.16788}, primaryclass = {cs.LG}, author = {Zimmer, Max and Spiegel, Christoph and Pokutta, Sebastian}, title = {Sparse Model Soups: A Recipe for Improved Pruning Via Model Averaging}, date = {2023-06-29} }
- Mundinger, K., Pokutta, S., Spiegel, C., and Zimmer, M. (2024, April 8). Extending the Continuum of Six-Colorings. Proceedings of the Discrete Mathematics Days.
[URL]
[arXiv]
[BibTeX]
@inproceedings{2024_MundingerPokuttaSpiegelZimmer_SixcoloringsExpansion:1, year = {2024}, booktitle = {Proceedings of the Discrete Mathematics Days}, url = {https://dmd2024.web.uah.es/files/abstracts/paper_27.pdf}, archiveprefix = {arXiv}, eprint = {2404.05509}, primaryclass = {math.CO}, author = {Mundinger, Konrad and Pokutta, Sebastian and Spiegel, Christoph and Zimmer, Max}, title = {Extending the Continuum of Six-Colorings}, date = {2024-04-08} }
- Pauls, J., Zimmer, M., Kelly, U. M., Schwartz, M., Saatchi, S., Ciais, P., Pokutta, S., Brandt, M., and Gieseke, F. (2024, June 3). Estimating Canopy Height at Scale. Proceedings of the International Conference on Machine Learning.
[arXiv]
[code]
[BibTeX]
@inproceedings{2024_PaulsEtAl_Canopyheightestimation, year = {2024}, booktitle = {Proceedings of the International Conference on Machine Learning}, archiveprefix = {arXiv}, eprint = {2406.01076}, primaryclass = {cs.CV}, author = {Pauls, Jan and Zimmer, Max and Kelly, Una M and Schwartz, Martin and Saatchi, Sassan and Ciais, Philippe and Pokutta, Sebastian and Brandt, Martin and Gieseke, Fabian}, title = {Estimating Canopy Height at Scale}, code = {https://github.com/AI4Forest/Global-Canopy-Height-Map}, date = {2024-06-03} }
- Zimmer, M., Spiegel, C., and Pokutta, S. (2021, November 1). How I Learned to Stop Worrying and Love Retraining. Proceedings of the International Conference on Learning Representations.
[URL]
[arXiv]
[code]
[BibTeX]
@inproceedings{2021_ZimmerSpiegelPokutta_Retrainingpruning, year = {2023}, booktitle = {Proceedings of the International Conference on Learning Representations}, url = {https://iclr.cc/virtual/2023/poster/10914}, archiveprefix = {arXiv}, eprint = {2111.00843}, primaryclass = {cs.LG}, author = {Zimmer, Max and Spiegel, Christoph and Pokutta, Sebastian}, title = {How I Learned to Stop Worrying and Love Retraining}, code = {https://github.com/ZIB-IOL/BIMP}, date = {2021-11-01} }
- Chmiela, A., Gleixner, A., Lichocki, P., and Pokutta, S. (2023). Online Learning for Scheduling MIP Heuristics. Proceedings of the International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research, 114â123.
DOI: 10.1007/978-3-031-33271-5_8
[arXiv]
[BibTeX]
@inproceedings{2023_ChmielaGleixnerLichockiPokutta_Onlinelearningscheduling, year = {2023}, date = {2023-05-23}, booktitle = {Proceedings of the International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research}, pages = {114-123}, doi = {10.1007/978-3-031-33271-5_8}, archiveprefix = {arXiv}, eprint = {2304.03755}, primaryclass = {math.OC}, author = {Chmiela, Antonia and Gleixner, Ambros and Lichocki, Pawel and Pokutta, Sebastian}, title = {Online Learning for Scheduling MIP Heuristics} }
- Mexi, G., Besançon, M., Bolusani, S., Chmiela, A., Hoen, A., and Gleixner, A. (2023, July 7). Scylla: a Matrix-free Fix-propagate-and-project Heuristic for Mixed-integer Optimization. Proceedings of the Conference of the Society for Operations Research in Germany.
[arXiv]
[BibTeX]
@inproceedings{2023_MexiEtAl_Scyllaheuristic, year = {2023}, booktitle = {Proceedings of the Conference of the Society for Operations Research in Germany}, archiveprefix = {arXiv}, eprint = {2307.03466}, primaryclass = {math.OC}, author = {Mexi, Gioni and Besançon, Mathieu and Bolusani, Suresh and Chmiela, Antonia and Hoen, Alexander and Gleixner, Ambros}, title = {Scylla: a Matrix-free Fix-propagate-and-project Heuristic for Mixed-integer Optimization}, date = {2023-07-07} }
- Thuerck, D., Sofranac, B., Pfetsch, M., and Pokutta, S. (2023, May 20). Learning Cuts Via Enumeration Oracles. Proceedings of the Conference on Neural Information Processing Systems.
[arXiv]
[BibTeX]
@inproceedings{2023_ThuerckSofranacPfetschPokutta_Learningcutsenumeration, year = {2023}, booktitle = {Proceedings of the Conference on Neural Information Processing Systems}, archiveprefix = {arXiv}, eprint = {2305.12197}, primaryclass = {math.OC}, author = {Thuerck, Daniel and Sofranac, Boro and Pfetsch, Marc and Pokutta, Sebastian}, title = {Learning Cuts Via Enumeration Oracles}, date = {2023-05-20} }
- Macdonald, J., Besançon, M., and Pokutta, S. (2021, October 15). Interpretable Neural Networks with Frank-Wolfe: Sparse Relevance Maps and Relevance Orderings. Proceedings of the International Conference on Machine Learning.
[arXiv]
[poster]
[video]
[BibTeX]
@inproceedings{2021_MacdonaldBesanconPokutta_Interpretableneuralnetworks, year = {2022}, booktitle = {Proceedings of the International Conference on Machine Learning}, archiveprefix = {arXiv}, eprint = {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}, date = {2021-10-15} }
- Tsuji, K., Tanaka, K., and Pokutta, S. (2021, October 4). Pairwise Conditional Gradients without Swap Steps and Sparser Kernel Herding. Proceedings of the International Conference on Machine Learning.
[arXiv]
[summary]
[slides]
[code]
[video]
[BibTeX]
@inproceedings{2021_TsujiTanakaPokutta_Pairwiseconditionalgradients, year = {2022}, booktitle = {Proceedings of the International Conference on Machine Learning}, archiveprefix = {arXiv}, eprint = {2110.12650}, primaryclass = {cond-mat.quant-gas}, author = {Tsuji, Kazuma and Tanaka, Ken'ichiro and Pokutta, Sebastian}, title = {Pairwise Conditional Gradients without Swap Steps and Sparser Kernel Herding}, code = {https://github.com/ZIB-IOL/FrankWolfe.jl}, slides = {https://pokutta.com/slides/20220624_ICML2022_BPCG.pdf}, summary = {https://pokutta.com/blog/research/2022/05/21/bpcg-abstract.html}, video = {https://slideslive.com/38983561}, date = {2021-10-04} }
- WĂ€ldchen, S., Huber, F., and Pokutta, S. (2022, February 23). Training Characteristic Functions with Reinforcement Learning: XAI-methods Play Connect Four. Proceedings of the International Conference on Machine Learning.
[arXiv]
[poster]
[video]
[BibTeX]
@inproceedings{2022_WaeldchenHuberPokutta_CharacteristicfunctionsReinforcementlearning, year = {2022}, booktitle = {Proceedings of the International Conference on Machine Learning}, archiveprefix = {arXiv}, eprint = {2202.11797}, primaryclass = {cs.LG}, author = {WĂ€ldchen, Stephan and Huber, Felix and Pokutta, Sebastian}, title = {Training Characteristic Functions with Reinforcement Learning: XAI-methods Play Connect Four}, poster = {https://pokutta.com/slides/20220712_icml_poster_conn4.pdf}, video = {https://slideslive.com/38983111}, date = {2022-02-23} }
- Chmiela, A., Khalil, E. B., Gleixner, A., Lodi, A., and Pokutta, S. (2021). Learning to Schedule Heuristics in Branch-and-bound. Proceedings of the Conference on Neural Information Processing Systems, 34, 24235â24246.
[URL]
[arXiv]
[poster]
[BibTeX]
@inproceedings{2021_ChmielaEtAl_Heuristicscheduling, year = {2021}, booktitle = {Proceedings of the Conference on Neural Information Processing Systems}, month = mar, volume = {34}, pages = {24235â24246}, url = {https://proceedings.neurips.cc/paper_files/paper/2021/file/cb7c403aa312160380010ee3dd4bfc53-Paper.pdf}, archiveprefix = {arXiv}, eprint = {2103.10294}, primaryclass = {cs.LG}, 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}, poster = {https://pokutta.com/slides/20211120_poster_NeurIPS21_learningheuristics.pdf}, date = {2021-03-18} }
- Ziemke, T., Sering, L., Vargas Koch, L., Zimmer, M., Nagel, K., and Skutella, M. (2020). Flows Over Time As Continuous Limits of Packet-based Network Simulations. Proceedings of the EURO Working Group on Transportation Meeting.
[BibTeX]
@inproceedings{2021_ZiemkeEtAl_Flowscontinuouslimits:1, year = {2020}, booktitle = {Proceedings of the EURO Working Group on Transportation Meeting}, author = {Ziemke, Theresa and Sering, Leon and Vargas Koch, Laura and Zimmer, Max and Nagel, Kai and Skutella, Martin}, title = {Flows Over Time As Continuous Limits of Packet-based Network Simulations} }
- Arumugam, K., Kadampot, I., Tahmasbi, M., Shah, S., Bloch, M., and Pokutta, S. (2017). Modulation Recognition Using Side Information and Hybrid Learning. Proceedings of the IEEE DySPAN.
[BibTeX]
@inproceedings{2017_ArumugamKadampotTahmasbiShahBlochPokutta_ModulationRecognition, year = {2017}, booktitle = {Proceedings of the IEEE DySPAN}, author = {Arumugam, K. and Kadampot, I. and Tahmasbi, M. and Shah, S. and Bloch, M. and Pokutta, Sebastian}, title = {Modulation Recognition Using Side Information and Hybrid Learning} }
- Roy, A., Xu, H., and Pokutta, S. (2017, June 15). Reinforcement Learning Under Model Mismatch. Proceedings of the Conference on Neural Information Processing Systems.
[arXiv]
[BibTeX]
@inproceedings{2017_RoyXuPokutta_ReinforcementLearning, year = {2017}, booktitle = {Proceedings of the Conference on Neural Information Processing Systems}, archiveprefix = {arXiv}, eprint = {1706.04711}, primaryclass = {cs.LG}, author = {Roy, Aurko and Xu, Huan and Pokutta, Sebastian}, title = {Reinforcement Learning Under Model Mismatch}, date = {2017-06-15} }
Full articles
- Mundinger, K., Pokutta, S., Spiegel, C., and Zimmer, M. (2024). Extending the Continuum of Six-Colorings. Geombinatorics Quarterly, XXXIV.
[URL]
[arXiv]
[BibTeX]
@article{2024_MundingerPokuttaSpiegelZimmer_SixcoloringsExpansion, year = {2024}, journal = {Geombinatorics Quarterly}, date = {2024-07-01}, volume = {XXXIV}, url = {https://geombina.uccs.edu/past-issues/volume-xxxiv}, archiveprefix = {arXiv}, eprint = {2404.05509}, primaryclass = {math.CO}, author = {Mundinger, Konrad and Pokutta, Sebastian and Spiegel, Christoph and Zimmer, Max}, title = {Extending the Continuum of Six-Colorings} }
- VuâHan, T. L., Sunkara, V., BermudezâSchettino, R., Schwechten, J., Runge, R., Perka, C., Winkler, T., Pokutta, S., WeiĂ, C., and Pumberger, M. (2024). Feature Engineering for the Prediction of Scoliosis in 5qâSpinal Muscular Atrophy. Journal of Cachexia, Sarcopenia and Muscle.
DOI: 10.1002/jcsm.13599
[URL]
[BibTeX]
@article{2024_Pokutta_PredictionScoliosis, year = {2024}, journal = {Journal of Cachexia, Sarcopenia and Muscle}, date = {2024-12-05}, doi = {10.1002/jcsm.13599}, url = {https://onlinelibrary.wiley.com/doi/pdf/10.1002/jcsm.13599}, author = {VuâHan, TuâLan and Sunkara, Vikram and BermudezâSchettino, Rodrigo and Schwechten, Jakob and Runge, Robin and Perka, Carsten and Winkler, Tobias and Pokutta, Sebastian and WeiĂ, Claudia and Pumberger, Matthias}, title = {Feature Engineering for the Prediction of Scoliosis in 5qâSpinal Muscular Atrophy} }
- Kevin-Martin, A., BĂ€rmann, A., Braun, K., Liers, F., Pokutta, S., Schneider, O., Sharma, K., and Tschuppik, S. (2023). Data-driven Distributionally Robust Optimization Over Time. INFORMS Journal on Optimization, 5(4), 376â394.
DOI: 10.1287/ijoo.2023.0091
[URL]
[arXiv]
[BibTeX]
@article{2023_KevinEtAl_Datadrivendistributionallyrobust, year = {2023}, journal = {INFORMS Journal on Optimization}, date = {2023-10-01}, month = oct, volume = {5}, number = {4}, pages = {376-394}, doi = {10.1287/ijoo.2023.0091}, url = {https://pubsonline.informs.org/doi/pdf/10.1287/ijoo.2023.0091}, archiveprefix = {arXiv}, eprint = {2304.05377}, primaryclass = {math.OC}, author = {Kevin-Martin, Aigner and BĂ€rmann, Andreas and Braun, Kristin and Liers, Frauke and Pokutta, Sebastian and Schneider, Oskar and Sharma, Kartikey and Tschuppik, Sebastian}, title = {Data-driven Distributionally Robust Optimization Over Time} }
- Bienstock, D., Muñoz, G., and Pokutta, S. (2023). Principled Deep Neural Network Training Through Linear Programming. Discrete Optimization, 49.
DOI: 10.1016/j.disopt.2023.100795
[arXiv]
[summary]
[BibTeX]
@article{2018_DanielGonzaloPokutta_Polyhedraltraining, year = {2023}, journal = {Discrete Optimization}, date = {2023-08-01}, month = aug, volume = {49}, doi = {10.1016/j.disopt.2023.100795}, archiveprefix = {arXiv}, eprint = {1810.03218}, primaryclass = {cs.LG}, author = {Bienstock, Daniel and Muñoz, Gonzalo and Pokutta, Sebastian}, title = {Principled Deep Neural Network Training Through Linear Programming}, summary = {https://www.pokutta.com/blog/research/2018/10/12/DNN-learning-lp-abstract.html} }
- Kruser, J., Sharma, K., Holl, J., and Nohadani, O. (2023). Identifying Patterns of Medical Intervention in Acute Respiratory Failure: A Retrospective Observational Study. Critical Care Explorations.
[BibTeX]
@article{2023_KruserSharmaHollNohadani_Medicalinterventionpatterns, year = {2023}, journal = {Critical Care Explorations}, date = {2023-09-30}, author = {Kruser, Jacqueline and Sharma, Kartikey and Holl, Jane and Nohadani, Omid}, title = {Identifying Patterns of Medical Intervention in Acute Respiratory Failure: A Retrospective Observational Study} }
- Kossen, T., Hirzel, M., Madai, V. I., Boenisch, F., Hennemuth, A., Hildebrand, K., Pokutta, S., Sharma, K., Hilbert, A., Sobesky, J., Galinovic, I., Khalil, A. A., Fiebach, J. B., and Frey, D. (2022). Towards Sharing Brain Images: Differentially Private TOF-MRA Images with Segmentation Labels Using Generative Adversarial Networks. Frontiers in Artificial Intelligence.
DOI: 10.3389/frai.2022.813842
[BibTeX]
@article{2022_KossenEtAl_Brainimagesharing, year = {2022}, journal = {Frontiers in Artificial Intelligence}, date = {2022-05-02}, doi = {10.3389/frai.2022.813842}, author = {Kossen, Tabea and Hirzel, Manuel and Madai, Vince I. and Boenisch, Franziska and Hennemuth, Anja and Hildebrand, Kristian and Pokutta, Sebastian and Sharma, Kartikey and Hilbert, Adam and Sobesky, Jan and Galinovic, Ivana and Khalil, Ahmed A. and Fiebach, Jochen B. and Frey, Dietmar}, title = {Towards Sharing Brain Images: Differentially Private TOF-MRA Images with Segmentation Labels Using Generative Adversarial Networks} }
- Nohadani, O., and Sharma, K. (2022). Optimization Under Connected Uncertainty. INFORMS Journal on Optimization.
DOI: 10.1287/ijoo.2021.0067
[arXiv]
[BibTeX]
@article{2022_NohadaniSharma_Connecteduncertainty, year = {2022}, journal = {INFORMS Journal on Optimization}, date = {2022-01-24}, doi = {10.1287/ijoo.2021.0067}, archiveprefix = {arXiv}, eprint = {2202.10602}, primaryclass = {math.OC}, author = {Nohadani, Omid and Sharma, Kartikey}, title = {Optimization Under Connected Uncertainty} }
- Pokutta, S. (2021). Mathematik, Machine Learning Und Artificial Intelligence. Mitteilungen Der DMV.
[URL]
[BibTeX]
@article{2021_Pokutta_MathematikMachineLearning, year = {2021}, journal = {Mitteilungen der DMV}, month = jan, url = {https://app.box.com/s/8u6fsucm78e1shyyzoghxar40f7xk6ni}, author = {Pokutta, Sebastian}, title = {Mathematik, Machine Learning Und Artificial Intelligence} }
- Ziemke, T., Sering, L., Vargas Koch, L., Zimmer, M., Nagel, K., and Skutella, M. (2021). Flows Over Time As Continuous Limits of Packet-based Network Simulations. Transportation Research Procedia, 52, 123â130.
DOI: 10.1016/j.trpro.2021.01.014
[URL]
[BibTeX]
@article{2021_ZiemkeEtAl_Flowscontinuouslimits, year = {2021}, journal = {Transportation Research Procedia}, volume = {52}, pages = {123-130}, doi = {10.1016/j.trpro.2021.01.014}, url = {https://sciencedirect.com/science/article/pii/S2352146521000284}, author = {Ziemke, Theresa and Sering, Leon and Vargas Koch, Laura and Zimmer, Max and Nagel, Kai and Skutella, Martin}, title = {Flows Over Time As Continuous Limits of Packet-based Network Simulations} }
- Klus, S., and GelĂ, P. (2019). Tensor-based Algorithms for Image Classification. Algorithms, 12(11), 240.
DOI: 10.3390/a12110240
[URL]
[arXiv]
[BibTeX]
@article{2019_KlusGelss_Tensoralgorithmsclassification, year = {2019}, journal = {Algorithms}, date = {2019-11-09}, volume = {12}, number = {11}, pages = {240}, doi = {10.3390/a12110240}, url = {https://mdpi.com/1999-4893/12/11/240}, archiveprefix = {arXiv}, eprint = {1910.02150}, primaryclass = {cs.LG}, author = {Klus, Stefan and GelĂ, Patrick}, title = {Tensor-based Algorithms for Image Classification} }
đŹ Projects
Ongoing Projects
Preserving global vegetation is crucial for addressing and mitigating climate change. Accurate, up-to-date forest health data is essential. AI4Forest aims to develop advanced AI methods to monitor forests using satellite imagery, including radar and optical data. The project will create scalable techniques for detailed, high-resolution maps of the globe, e.g., to monitor canopy height, biomass, and to track forest disturbances.
In this project, we study domain decomposition approaches for optimal control in gas transport networks. Our goal is to couple space-time-domain decomposition with machine learning and mixed-integer programming. We will develop NeTI (Network Tearing and Interconnection), a data-driven and physics-informed algorithm combining mixed-integer nonlinear programming, surrogate model learning, and graph decomposition strategies.
Existing approaches for interpreting Neural Network classifiers that highlight features relevant for a decision are based solely on heuristics. We introduce a theory that allows us to bound the quality of the features without assumptions on the classifier model by relating classification to Interactive Proof Systems.
Completed Projects
Heuristics play a crucial role in exact solvers for Mixed Integer Programming (MIP). However, the question of how to manage multiple MIP heuristics in a solver has not received sufficient attention. This project addresses the strategic management of primal heuristics in MIP solvers, aiming to replace static, hard-coded rules with dynamic, self-improving procedures.
The performance of modern mixed-integer program solvers is highly dependent on a number of interdependent individual components. Using tools from machine learning, we intend to develop an integrated framework that is able to capture interactions of individual decisions made in these components with the ultimate goal to improve performance.
Training artificial neural networks is a key optimization task in deep learning. To improve generalization, robustness, and explainability, we aim to compute globally optimal solutions. We will use integer programming methods, exploiting mixed-integer nonlinear programming and enhancing solving techniques like spatial branch-and-cut. Additionally, we'll leverage symmetry to reduce computational burden and ensure symmetry in solutions, and incorporate true sparsity using a mixed-integer nonlinear programming framework.