Profiles2020-2026
Sebastian Pokutta
Technische Universität Berlin
21Papers
17Last author
35Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers21
- Fast Frank-Wolfe Algorithms with Adaptive Bregman Step-Size for Weakly Convex FunctionsAuthor 2 of 3ICLR 2026 · 10 citations
- Beyond Short Steps in Frank-Wolfe AlgorithmsAuthor 2 of 2ICLR 2026 · 5 citations
- RECON: Robust symmetry discovery via Explicit Canonical Orientation NormalizationAuthor 4 of 4ICLR 2026 · 1 citation
- Efficient Quadratic Corrections for Frank-Wolfe AlgorithmsAuthor 5 of 5NeurIPS 2025 · 6 citations
- GSE: Group-wise Sparse and Explainable Adversarial AttacksAuthor 3 of 3ICLR 2025
- On the Byzantine-Resilience of Distillation-Based Federated LearningAuthor 3 of 3ICLR 2025
- Secant Line Search for Frank-Wolfe AlgorithmsAuthor 2 of 4ICML 2025
- The Good, the Bad and the Ugly: Meta-Analysis of Watermarks, Transferable Attacks and Adversarial DefensesAuthor 5 of 5NeurIPS 2025
- Sparse Model Soups: A Recipe for Improved Pruning via Model AveragingAuthor 3 of 3ICLR 2024 · 22 citations
- Learning Cuts via Enumeration OraclesAuthor 5 of 6NeurIPS 2023 · 7 citations
- Fully Computer-Assisted Proofs in Extremal CombinatoricsAuthor 2 of 4AAAI 2023 · 4 citations
- Approximate Vanishing Ideal Computations at ScaleAuthor 3 of 3ICLR 2023
- How I Learned to Stop Worrying and Love RetrainingAuthor 3 of 3ICLR 2023
- Interpretable Neural Networks with Frank-Wolfe: Sparse Relevance Maps and Relevance OrderingsAuthor 3 of 3ICML 2022 · 13 citations
- Training Characteristic Functions with Reinforcement Learning: XAI-methods play Connect FourAuthor 3 of 3ICML 2022 · 9 citations
- Fast Algorithms for Packing Proportional Fairness and its DualAuthor 3 of 3NeurIPS 2022 · 4 citations
- Parameter-free Locally Accelerated Conditional GradientsAuthor 5 of 5ICML 2021 · 9 citations
- Boosting Frank-Wolfe by Chasing GradientsAuthor 2 of 2ICML 2020 · 32 citations
- On the Unreasonable Effectiveness of the Greedy Algorithm: Greedy Adapts to SharpnessAuthor 3 of 3ICML 2020 · 12 citations
- Walking in the Shadow: A New Perspective on Descent Directions for Constrained MinimizationAuthor 3 of 3NeurIPS 2020 · 8 citations
- IPBoost - Non-Convex Boosting via Integer ProgrammingAuthor 2 of 2ICML 2020 · 6 citations
