Sebastian Pokutta

Sebastian Pokutta

Artificial Intelligence, Optimization, and Machine Learning. Professor @TUBerlin, Vice President @ZuseInstitute



Prof. Dr. Sebastian Pokutta

Vice President and Division Head
Mathematical Algorithmic Intelligence
AI in Society, Science, and Technology (AIS²T)
Zuse Institute Berlin (ZIB)

Professor for Optimization and Machine Learning
Institute of Mathematics
Electrical Engineering and Computer Science (courtesy)
Technische Universität Berlin

Research Lab. My group is interested in Artificial Intelligence, Optimization, and Machine Learning. We develop new methodologies (e.g., new optimization and learning algorithms), work on combining learning and decision-making, as well as design AI Systems for real-world deployment in various application contexts. [more]

(Informal) TL;DR. We use computers to learn from data and make better decisions.

Prospective Students. If you are interested in working in our group or writing your MS/BS thesis please only use the email applications-aisst@zib.de.

Recent Papers.

  1. Wirth, E., Kera, H., and Pokutta, S. (2022). Approximate Vanishing Ideal Computations at Scale. Preprint. [arXiv]
  2. Parczyk, O., Pokutta, S., Spiegel, C., and Szabó, T. (2022). New Ramsey Multiplicity Bounds and Search Heuristics. Preprint. [arXiv] [slides]
  3. Wäldchen, S., Sharma, K., Zimmer, M., and Pokutta, S. (2022). Merlin-Arthur Classifiers: Formal Interpretability with Interactive Black Boxes. Preprint. [arXiv]
  4. Deza, A., Pokutta, S., and Pournin, L. (2022). The complexity of geometric scaling. Preprint. [arXiv]
  5. Wäldchen, S., Huber, F., and Pokutta, S. (2022). Training Characteristic Functions with Reinforcement Learning: XAI-methods play Connect Four. To Appear in Proceedings of ICML. [arXiv] [poster]
  6. Zimmer, M., Spiegel, C., and Pokutta, S. (2022). Compression-aware Training of Neural Networks using Frank-Wolfe. Preprint. [arXiv]
  7. Gelß, P., Klus, S., Shakibaei, Z., and Pokutta, S. (2022). Low-rank tensor decompositions of quantum circuits. Preprint. [arXiv]
  8. Tsuji, K., Tanaka, K., and Pokutta, S. (2022). Pairwise Conditional Gradients without Swap Steps and Sparser Kernel Herding. To Appear in Proceedings of ICML. [arXiv] [summary] [slides] [code]
  9. Macdonald, J., Besançon, M., and Pokutta, S. (2022). Interpretable Neural Networks with Frank-Wolfe: Sparse Relevance Maps and Relevance Orderings. To Appear in Proceedings of ICML. [arXiv] [poster]
  10. Kossen, T., Hirzel, M. A., 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. [PDF]
  11. Gasse, M., Cappart, Q., Charfreitag, J., Charlin, L., Chételat, D., Chmiela, A., Dumouchelle, J., Gleixner, A., Kazachkov, A. M., Khalil, E., Lichocki, P., Lodi, A., Lubin, M., Maddison, C. J., Morris, C., Papageorgiou, D. J., Parjadis, A., Pokutta, S., Prouvost, A., … Kun, M. (2022). The Machine Learning for Combinatorial Optimization Competition (ML4CO): Results and Insights. Preprint. [arXiv]
  12. Hunkenschröder, C., Pokutta, S., and Weismantel, R. (2022). Optimizing a low-dimensional convex function over a high-dimensional cube. Preprint. [arXiv]
  13. Kerdreux, T., Scieur, D., d’Aspremont, A., and Pokutta, S. (2022). Strong Convexity of Feasible Sets in Riemannian Manifolds. Preprint.
  14. Wirth, E., Kerdreux, T., and Pokutta, S. (2022). Acceleration of Frank-Wolfe algorithms with open loop step-sizes. Preprint. [arXiv]
  15. Wirth, E., and Pokutta, S. (2022). Conditional Gradients for the Approximately Vanishing Ideal. To Appear in Proceedings of AISTATS. [arXiv] [summary] [poster] [code]
  16. Zimmer, M., Spiegel, C., and Pokutta, S. (2021). How I Learned to Stop Worrying and Love Retraining. Preprint. [arXiv] [code]
  17. Criado, F., Martinez-Rubio, D., and Pokutta, S. (2021). Fast Algorithms for Packing Proportional Fairness and its Dual. Preprint. [arXiv] [poster]
  18. Roux, C., Wirth, E., Pokutta, S., and Kerdreux, T. (2021). Efficient Online-Bandit Strategies for Minimax Learning Problems. Preprint. [arXiv]
  19. Kerdreux, T., d’Aspremont, A., and Pokutta, S. (2021). Local and Global Uniform Convexity Conditions. Preprint. [arXiv]
  20. Braun, G., and Pokutta, S. (2021). Dual Prices for Frank-Wolfe Algorithms. Preprint. [arXiv]

Select Recent Talks and Teaching.

Recent Blog Posts.

News.

  • 06/2022: Symposium on Theory of Computing (STOC) Test of Time award (10 years) for “Linear vs. Semidefinite Extended Formulations: Exponential Separation and Strong Lower Bounds”, S. Fiorini, S. Massar, S. Pokutta, H.R. Tiwary, R. de Wolf from 2012.
  • 06/2022: 6th RIKEN-IMI-ISM-ZIB-MODAL-NHR Workshop on Advances in Classical and Quantum Algorithms for Optimization and Machine Learning in Japan. [link]
  • 06/2022: New ZIB videos available. [youtube channel]. (german only)
  • 06/2022: Interview on using AI to combat and mitigate climate change (German) [article] [magazine]
  • 10/2021: Math+ Cluster presentation at the Humboldt Forum “Mit Mathematik die Welt verbessern?” (German) [video]