AIConfPaper
← Search

Hubert Baniecki

5 accepted papers

2026

Functional Decomposition and Shapley Interactions for Interpreting Survival Models

Sophie Hanna Langbein, Hubert Baniecki, Fabian Fumagalli, Niklas Koenen, Marvin N. Wright, Julia Herbinger
ICML 2026poster

Hazard and survival functions are natural, interpretable targets in time-to-event prediction, but their inherent non-additivity fundamentally limits standard additive explanation methods. We introduce Survival Functional Decomposition (SurvFD), a principled approach for analyzing feature interaction…

Cited by 0Source ↗Scholar ↗
2025

Efficient and Accurate Explanation Estimation with Distribution Compression

Hubert Baniecki, Giuseppe Casalicchio, Bernd Bischl, Przemyslaw Biecek
ICLR 2025spotlight

We discover a theoretical connection between explanation estimation and distribution compression that significantly improves the approximation of feature attributions, importance, and effects. While the exact computation of various machine learning explanations requires numerous model inferences and…

Cited by 2Source ↗PDF ↗code ↗Scholar ↗
2025

Explaining Similarity in Vision-Language Encoders with Weighted Banzhaf Interactions

Hubert Baniecki, Maximilian Muschalik, Fabian Fumagalli, Barbara Hammer, Eyke Hüllermeier, Przemyslaw Biecek
NeurIPS 2025poster

Language-image pre-training (LIP) enables the development of vision-language models capable of zero-shot classification, localization, multimodal retrieval, and semantic understanding. Various explanation methods have been proposed to visualize the importance of input image-text pairs on the model's…

Cited by 0Source ↗Scholar ↗
2025

Interpreting CLIP with Hierarchical Sparse Autoencoders

Vladimir Zaigrajew, Hubert Baniecki, Przemyslaw Biecek
ICML 2025poster

Sparse autoencoders (SAEs) are useful for detecting and steering interpretable features in neural networks, with particular potential for understanding complex multimodal representations. Given their ability to uncover interpretable features, SAEs are particularly valuable for analyzing vision-langu…

Cited by 0Source ↗PDF ↗code ↗Scholar ↗
2024

shapiq: Shapley Interactions for Machine Learning

Maximilian Muschalik, Hubert Baniecki, Fabian Fumagalli, Patrick Kolpaczki, Barbara Hammer, Eyke Hüllermeier
NeurIPS 2024poster

Originally rooted in game theory, the Shapley Value (SV) has recently become an important tool in machine learning research. Perhaps most notably, it is used for feature attribution and data valuation in explainable artificial intelligence. Shapley Interactions (SIs) naturally extend the SV and addr…

Cited by 11Source ↗PDF ↗code ↗Scholar ↗