publications

2026

  1. Preprint
    An Odd Estimator for Shapley Values
    Fabian Fumagalli, Landon Butler, Justin Singh Kang, Kannan Ramchandran, and R. Teal Witter
    2026
    PDF
  2. ICLR 2026
    PolySHAP: Extending KernelSHAP with Interaction-Informed Polynomial Regression
    Fabian Fumagalli, R. Teal Witter, and Christopher Musco
    In Proceedings of the International Conference on Learning Representations (ICLR), 2026
    PDF
  3. AISTATS 2026
    GRANITE: A Generalized Regional Framework for Identifying Agreement in Feature-Based Explanations
    Julia Herbinger, Gabriel Laberge, Maximilian Muschalik, Yann Pequignot, Marvin N. Wright, and Fabian Fumagalli
    In Proceedings of the International Conference on Artificial Intelligence and Statistics (AISTATS), 2026
    PDF
  4. AAAI 2026 (Oral)
    HyperSHAP: Shapley Values and Interactions for Hyperparameter Importance
    Marcel Wever, Maximilian Muschalik, Fabian Fumagalli, and Marius Lindauer
    In Fortieth AAAI Conference on Artificial Intelligence, (AAAI), 2026
    PDF

2025

  1. NeurIPS 2025
    Explaining Similarity in Vision-Language Encoders with Weighted Banzhaf Interactions
    Hubert Baniecki, Maximilian Muschalik, Fabian Fumagalli, Barbara Hammer, Eyke Hüllermeier, and Przemyslaw Biecek
    In Proceedings of Advances in Neural Information Processing Systems (NeurIPS), 2025
    PDF
  2. EMNLP 2025
    Investigating the Impact of Conceptual Metaphors on LLM-based NLI through Shapley Interactions
    Meghdut Sengupta, Maximilian Muschalik, Fabian Fumagalli, Barbara Hammer, Eyke Hüllermeier, Debanjan Ghosh, and Henning Wachsmuth
    In Findings of the Association for Computational Linguistics: EMNLP 2025, 2025
    PDF
  3. NAACL 2025 (Oral)
    Adaptive Prompting: Ad-hoc Prompt Composition for Social Bias Detection
    Maximilian Spliethöver, Tim Knebler, Fabian Fumagalli, Maximilian Muschalik, Barbara Eva Hammer, Eyke Hüllermeier, and Henning Wachsmuth
    In The 2025 Annual Conference of the Nations of the Americas Chapter of the ACL, 2025
    PDF
  4. PhD Thesis
    Explainable AI: A Unified Approach Based on Cooperative Game Theory
    Fabian Fumagalli
    Bielefeld University, 2025
    PDF
  5. ICLR 2025
    Exact Computation of Any-Order Shapley Interactions for Graph Neural Networks
    Maximilian Muschalik*, Fabian Fumagalli*, Paolo Frazzetto, Janine Strotherm, Luca Hermes, Alessandro Sperduti, Eyke Hüllermeier, and Barbara Hammer
    In The Thirteenth International Conference on Learning Representations, 2025
    PDF
  6. AISTATS 2025
    Unifying Feature-Based Explanations with Functional ANOVA and Cooperative Game Theory
    Fabian Fumagalli, Maximilian Muschalik, Eyke Hüllermeier, Barbara Hammer, and Julia Herbinger
    In The 28th International Conference on Artificial Intelligence and Statistics, 2025
    PDF

2024

  1. NeurIPS 2024
    shapiq: Shapley Interactions for Machine Learning
    Maximilian Muschalik, Hubert Baniecki, Fabian Fumagalli, Patrick Kolpaczki, Barbara Hammer, and Eyke Hüllermeier
    In The Thirty-eight Conference on Neural Information Processing Systems (NeurIPS) Datasets and Benchmarks Track, 2024
    PDF Code Poster
  2. ICML 2024
    KernelSHAP-IQ: Weighted Least Square Optimization for Shapley Interactions
    Fabian Fumagalli, Maximilian Muschalik, Patrick Kolpaczki, Eyke Hüllermeier, and Barbara Hammer
    In Forty-first International Conference on Machine Learning (ICML), 2024
    PDF Poster
  3. AAAI 2024
    Beyond TreeSHAP: Efficient Computation of Any-Order Shapley Interactions for Tree Ensembles
    Maximilian Muschalik*, Fabian Fumagalli*, Barbara Hammer, and Eyke Hüllermeier
    In Thirty-Eighth AAAI Conference on Artificial Intelligence, (AAAI), 2024
    DOI PDF Poster
  4. AISTATS 2024
    SVARM-IQ: Efficient Approximation of Any-order Shapley Interactions through Stratification
    Patrick Kolpaczki, Maximilian Muschalik, Fabian Fumagalli, Barbara Hammer, and Eyke Hüllermeier
    In International Conference on Artificial Intelligence and Statistics (AISTATS), 2024
    PDF
  5. IJCNN 2024
    No learning rates needed: Introducing SALSA - Stable Armijo Line Search Adaptation
    Philip Kenneweg, Tristan Kenneweg, Fabian Fumagalli, and Barbara Hammer
    In International Joint Conference on Neural Networks, (IJCNN), 2024
    DOI PDF

2023

  1. NeurIPS 2023
    SHAP-IQ: Unified Approximation of any-order Shapley Interactions
    Fabian Fumagalli*, Maximilian Muschalik*, Patrick Kolpaczki, Eyke Hüllermeier, and Barbara Hammer
    In Advances in Neural Information Processing Systems 36 (NeurIPS), 2023
    PDF Poster
  2. MLJ
    Incremental Permutation Feature Importance (iPFI): Towards Online Explanations on Data Streams
    Fabian Fumagalli*, Maximilian Muschalik*, Eyke Hüllermeier, and Barbara Hammer
    Machine Learning, 2023
    DOI PDF Poster Slides
  3. xAI 2023
    iPDP: On Partial Dependence Plots in Dynamic Modeling Scenarios
    Maximilian Muschalik*, Fabian Fumagalli*, Rohit Jagtani, Barbara Hammer, and Eyke Hüllermeier
    In World Conference of Explainable Artificial Intelligence (xAI), 2023
    DOI PDF Slides
  4. ECML 2023
    iSAGE: An Incremental Version of SAGE for Online Explanation on Data Streams
    Maximilian Muschalik*, Fabian Fumagalli*, Barbara Hammer, and Eyke Hüllermeier
    In Machine Learning and Knowledge Discovery in Databases: Research Track - European Conference, (ECML PKDD), 2023
    DOI PDF Poster Slides
  5. ESANN 2023
    On Feature Removal for Explainability in Dynamic Environments
    Fabian Fumagalli*, Maximilian Muschalik*, Eyke Hüllermeier, and Barbara Hammer
    In 31st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN), 2023
    DOI Poster

2022

  1. KI Journal
    Agnostic Explanation of Model Change based on Feature Importance
    Maximilian Muschalik*, Fabian Fumagalli*, Barbara Hammer, and Eyke Hüllermeier
    Künstliche Intelligenz, 2022
    DOI PDF