@inproceedings{MartinezRubioRouxPokutta2024, author = {Mart{\´i}nez-Rubio, David and Roux, Christophe and Pokutta, Sebastian}, title = {Convergence and Trade-Offs in Riemannian Gradient Descent and Riemannian Proximal Point}, volume = {235}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {34920 -- 34948}, year = {2024}, abstract = {In this work, we analyze two of the most fundamental algorithms in geodesically convex optimization: Riemannian gradient descent and (possibly inexact) Riemannian proximal point. We quantify their rates of convergence and produce different variants with several trade-offs. Crucially, we show the iterates naturally stay in a ball around an optimizer, of radius depending on the initial distance and, in some cases, on the curvature. Previous works simply assumed bounded iterates, resulting in rates that were not fully quantified. We also provide an implementable inexact proximal point algorithm and prove several new useful properties of Riemannian proximal methods: they work when positive curvature is present, the proximal operator does not move points away from any optimizer, and we quantify the smoothness of its induced Moreau envelope. Further, we explore beyond our theory with empirical tests.}, language = {en} } @inproceedings{WaeldchenSharmaTuranetal.2024, author = {W{\"a}ldchen, Stephan and Sharma, Kartikey and Turan, Berkant and Zimmer, Max and Pokutta, Sebastian}, title = {Interpretability Guarantees with Merlin-Arthur Classifiers}, volume = {238}, booktitle = {Proceedings of The 27th International Conference on Artificial Intelligence and Statistics}, pages = {1963 -- 1971}, year = {2024}, abstract = {We propose an interactive multi-agent classifier that provides provable interpretability guarantees even for complex agents such as neural networks. These guarantees consist of lower bounds on the mutual information between selected features and the classification decision. Our results are inspired by the Merlin-Arthur protocol from Interactive Proof Systems and express these bounds in terms of measurable metrics such as soundness and completeness. Compared to existing interactive setups, we rely neither on optimal agents nor on the assumption that features are distributed independently. Instead, we use the relative strength of the agents as well as the new concept of Asymmetric Feature Correlation which captures the precise kind of correlations that make interpretability guarantees difficult. We evaluate our results on two small-scale datasets where high mutual information can be verified explicitly.}, language = {en} } @inproceedings{ZimmerSpiegelPokutta2024, author = {Zimmer, Max and Spiegel, Christoph and Pokutta, Sebastian}, title = {Sparse Model Soups}, booktitle = {12th International Conference on Learning Representations (ICLR 2024)}, publisher = {Curran Associates, Inc.}, isbn = {9781713898658}, year = {2024}, language = {en} } @inproceedings{TroppensBesanconWilkenetal.2025, author = {Troppens, Hannah and Besan{\c{c}}on, Mathieu and Wilken, St. Elmo and Pokutta, Sebastian}, title = {Mixed-Integer Optimization for Loopless Flux Distributions in Metabolic Networks}, volume = {338}, booktitle = {23rd International Symposium on Experimental Algorithms (SEA 2025)}, publisher = {Schloss Dagstuhl - Leibniz-Zentrum f{\"u}r Informatik}, doi = {10.4230/LIPIcs.SEA.2025.26}, pages = {26:1 -- 26:18}, year = {2025}, language = {en} } @inproceedings{GłuchTuranNagarajanetal.2025, author = {Głuch, 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}, booktitle = {1st Workshop on GenAI Watermarking, collocated with ICLR 2025}, year = {2025}, language = {en} } @inproceedings{BesanconPokuttaWirth2025, author = {Besan{\c{c}}on, Mathieu and Pokutta, Sebastian and Wirth, Elias}, title = {The Pivoting Framework: Frank-Wolfe Algorithms with Active Set Size Control}, volume = {258}, booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics}, pages = {271 -- 279}, year = {2025}, language = {en} } @inproceedings{MartinezRubioRouxCriscitielloetal.2025, author = {Mart{\´i}nez-Rubio, David and Roux, Christophe and Criscitiello, Christopher and Pokutta, Sebastian}, title = {Accelerated Methods for Riemannian Min-Max Optimization Ensuring Bounded Geometric Penalties}, volume = {258}, booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics}, pages = {280 -- 288}, year = {2025}, language = {en} } @inproceedings{SadikuWagnerPokutta2025, author = {Sadiku, Shpresim and Wagner, Moritz and Pokutta, Sebastian}, title = {GSE: Group-wise sparse and explainable adversarial attacks}, booktitle = {13th International Conference on Learning Representations (ICLR 2025)}, arxiv = {http://arxiv.org/abs/2311.17434}, year = {2025}, language = {en} } @inproceedings{RouxZimmerPokutta2025, author = {Roux, Christophe and Zimmer, Max and Pokutta, Sebastian}, title = {On the byzantine-resilience of distillation-based federated learning}, booktitle = {13th International Conference on Learning Representations (ICLR 2025)}, arxiv = {http://arxiv.org/abs/2402.12265}, year = {2025}, abstract = {Federated Learning (FL) algorithms using Knowledge Distillation (KD) have received increasing attention due to their favorable properties with respect to privacy, non-i.i.d. data and communication cost. These methods depart from transmitting model parameters and instead communicate information about a learning task by sharing predictions on a public dataset. In this work, we study the performance of such approaches in the byzantine setting, where a subset of the clients act in an adversarial manner aiming to disrupt the learning process. We show that KD-based FL algorithms are remarkably resilient and analyze how byzantine clients can influence the learning process. Based on these insights, we introduce two new byzantine attacks and demonstrate their ability to break existing byzantine-resilient methods. Additionally, we propose a novel defence method which enhances the byzantine resilience of KD-based FL algorithms. Finally, we provide a general framework to obfuscate attacks, making them significantly harder to detect, thereby improving their effectiveness.}, language = {en} } @inproceedings{SadikuWagnerNagarajanetal.2025, author = {Sadiku, Shpresim and Wagner, Moritz and Nagarajan, Sai Ganesh and Pokutta, Sebastian}, title = {S-CFE: Simple Counterfactual Explanations}, volume = {258}, booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics}, pages = {172 -- 180}, year = {2025}, language = {en} }