---
_id: '61190'
author:
- first_name: Meghdut
full_name: Sengupta, Meghdut
id: '99459'
last_name: Sengupta
- first_name: 'Maximilian '
full_name: 'Muschalik, Maximilian '
last_name: Muschalik
- first_name: Fabian
full_name: Fumagalli, Fabian
id: '93420'
last_name: Fumagalli
- first_name: Barbara
full_name: Hammer, Barbara
last_name: Hammer
- first_name: 'Eyke '
full_name: 'Hüllermeier, Eyke '
last_name: Hüllermeier
- first_name: Debanjan
full_name: Ghosh, Debanjan
last_name: Ghosh
- first_name: Henning
full_name: Wachsmuth, Henning
id: '3900'
last_name: Wachsmuth
citation:
ama: 'Sengupta M, Muschalik M, Fumagalli F, et al. Investigating the Impact of Conceptual
Metaphors on LLM-based NLI through Shapley Interactions. In: Accepted in Findings
. EMNLP ; 2025.'
apa: Sengupta, M., Muschalik, M., Fumagalli, F., Hammer, B., Hüllermeier, E., Ghosh,
D., & Wachsmuth, H. (2025). Investigating the Impact of Conceptual Metaphors
on LLM-based NLI through Shapley Interactions. Accepted in Findings . Empirical
Methods in Natural Language Processing (EMNLP 2025).
bibtex: '@inproceedings{Sengupta_Muschalik_Fumagalli_Hammer_Hüllermeier_Ghosh_Wachsmuth_2025,
title={Investigating the Impact of Conceptual Metaphors on LLM-based NLI through
Shapley Interactions}, booktitle={Accepted in Findings }, publisher={EMNLP },
author={Sengupta, Meghdut and Muschalik, Maximilian and Fumagalli, Fabian and
Hammer, Barbara and Hüllermeier, Eyke and Ghosh, Debanjan and Wachsmuth, Henning},
year={2025} }'
chicago: Sengupta, Meghdut, Maximilian Muschalik, Fabian Fumagalli, Barbara Hammer,
Eyke Hüllermeier, Debanjan Ghosh, and Henning Wachsmuth. “Investigating the Impact
of Conceptual Metaphors on LLM-Based NLI through Shapley Interactions.” In Accepted
in Findings . EMNLP , 2025.
ieee: M. Sengupta et al., “Investigating the Impact of Conceptual Metaphors
on LLM-based NLI through Shapley Interactions,” presented at the Empirical Methods
in Natural Language Processing (EMNLP 2025), 2025.
mla: Sengupta, Meghdut, et al. “Investigating the Impact of Conceptual Metaphors
on LLM-Based NLI through Shapley Interactions.” Accepted in Findings ,
EMNLP , 2025.
short: 'M. Sengupta, M. Muschalik, F. Fumagalli, B. Hammer, E. Hüllermeier, D. Ghosh,
H. Wachsmuth, in: Accepted in Findings , EMNLP , 2025.'
conference:
end_date: 2025-11-09
name: Empirical Methods in Natural Language Processing (EMNLP 2025)
start_date: 2025-11-04
date_created: 2025-09-11T09:27:10Z
date_updated: 2025-09-11T09:28:23Z
ddc:
- '000'
extern: '1'
file:
- access_level: closed
content_type: application/pdf
creator: meghdut
date_created: 2025-09-11T09:26:18Z
date_updated: 2025-09-11T09:26:18Z
file_id: '61194'
file_name: sengupta25-emnlp.pdf
file_size: 2189153
relation: main_file
success: 1
file_date_updated: 2025-09-11T09:26:18Z
has_accepted_license: '1'
language:
- iso: eng
project:
- _id: '127'
name: 'TRR 318; TP C04: Metaphern als Werkzeug des Erklärens'
publication: 'Accepted in Findings '
publisher: 'EMNLP '
status: public
title: Investigating the Impact of Conceptual Metaphors on LLM-based NLI through Shapley
Interactions
type: conference
user_id: '99459'
year: '2025'
...
---
_id: '59856'
abstract:
- lang: eng
text: Recent advances on instruction fine-tuning have led to the development of
various prompting techniques for large language models, such as explicit reasoning
steps. However, the success of techniques depends on various parameters, such
as the task, language model, and context provided. Finding an effective prompt
is, therefore, often a trial-and-error process. Most existing approaches to automatic
prompting aim to optimize individual techniques instead of compositions of techniques
and their dependence on the input. To fill this gap, we propose an adaptive prompting
approach that predicts the optimal prompt composition ad-hoc for a given input.
We apply our approach to social bias detection, a highly context-dependent task
that requires semantic understanding. We evaluate it with three large language
models on three datasets, comparing compositions to individual techniques and
other baselines. The results underline the importance of finding an effective
prompt composition. Our approach robustly ensures high detection performance,
and is best in several settings. Moreover, first experiments on other tasks support
its generalizability.
author:
- first_name: Maximilian
full_name: Spliethöver, Maximilian
id: '84035'
last_name: Spliethöver
orcid: 0000-0003-4364-1409
- first_name: Tim
full_name: Knebler, Tim
last_name: Knebler
- first_name: Fabian
full_name: Fumagalli, Fabian
id: '93420'
last_name: Fumagalli
- first_name: Maximilian
full_name: Muschalik, Maximilian
last_name: Muschalik
- first_name: Barbara
full_name: Hammer, Barbara
last_name: Hammer
- first_name: Eyke
full_name: Hüllermeier, Eyke
id: '48129'
last_name: Hüllermeier
- first_name: Henning
full_name: Wachsmuth, Henning
id: '3900'
last_name: Wachsmuth
citation:
ama: 'Spliethöver M, Knebler T, Fumagalli F, et al. Adaptive Prompting: Ad-hoc Prompt
Composition for Social Bias Detection. In: Chiruzzo L, Ritter A, Wang L, eds.
Proceedings of the 2025 Conference of the Nations of the Americas Chapter of
the Association for Computational Linguistics: Human Language Technologies (Volume
1: Long Papers). Association for Computational Linguistics; 2025:2421–2449.'
apa: 'Spliethöver, M., Knebler, T., Fumagalli, F., Muschalik, M., Hammer, B., Hüllermeier,
E., & Wachsmuth, H. (2025). Adaptive Prompting: Ad-hoc Prompt Composition
for Social Bias Detection. In L. Chiruzzo, A. Ritter, & L. Wang (Eds.), Proceedings
of the 2025 Conference of the Nations of the Americas Chapter of the Association
for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
(pp. 2421–2449). Association for Computational Linguistics.'
bibtex: '@inproceedings{Spliethöver_Knebler_Fumagalli_Muschalik_Hammer_Hüllermeier_Wachsmuth_2025,
place={Albuquerque, New Mexico}, title={Adaptive Prompting: Ad-hoc Prompt Composition
for Social Bias Detection}, booktitle={Proceedings of the 2025 Conference of the
Nations of the Americas Chapter of the Association for Computational Linguistics:
Human Language Technologies (Volume 1: Long Papers)}, publisher={Association for
Computational Linguistics}, author={Spliethöver, Maximilian and Knebler, Tim and
Fumagalli, Fabian and Muschalik, Maximilian and Hammer, Barbara and Hüllermeier,
Eyke and Wachsmuth, Henning}, editor={Chiruzzo, Luis and Ritter, Alan and Wang,
Lu}, year={2025}, pages={2421–2449} }'
chicago: 'Spliethöver, Maximilian, Tim Knebler, Fabian Fumagalli, Maximilian Muschalik,
Barbara Hammer, Eyke Hüllermeier, and Henning Wachsmuth. “Adaptive Prompting:
Ad-Hoc Prompt Composition for Social Bias Detection.” In Proceedings of the
2025 Conference of the Nations of the Americas Chapter of the Association for
Computational Linguistics: Human Language Technologies (Volume 1: Long Papers),
edited by Luis Chiruzzo, Alan Ritter, and Lu Wang, 2421–2449. Albuquerque, New
Mexico: Association for Computational Linguistics, 2025.'
ieee: 'M. Spliethöver et al., “Adaptive Prompting: Ad-hoc Prompt Composition
for Social Bias Detection,” in Proceedings of the 2025 Conference of the Nations
of the Americas Chapter of the Association for Computational Linguistics: Human
Language Technologies (Volume 1: Long Papers), 2025, pp. 2421–2449.'
mla: 'Spliethöver, Maximilian, et al. “Adaptive Prompting: Ad-Hoc Prompt Composition
for Social Bias Detection.” Proceedings of the 2025 Conference of the Nations
of the Americas Chapter of the Association for Computational Linguistics: Human
Language Technologies (Volume 1: Long Papers), edited by Luis Chiruzzo et
al., Association for Computational Linguistics, 2025, pp. 2421–2449.'
short: 'M. Spliethöver, T. Knebler, F. Fumagalli, M. Muschalik, B. Hammer, E. Hüllermeier,
H. Wachsmuth, in: L. Chiruzzo, A. Ritter, L. Wang (Eds.), Proceedings of the 2025
Conference of the Nations of the Americas Chapter of the Association for Computational
Linguistics: Human Language Technologies (Volume 1: Long Papers), Association
for Computational Linguistics, Albuquerque, New Mexico, 2025, pp. 2421–2449.'
date_created: 2025-05-10T12:37:45Z
date_updated: 2025-09-12T09:51:30Z
department:
- _id: '660'
editor:
- first_name: Luis
full_name: Chiruzzo, Luis
last_name: Chiruzzo
- first_name: Alan
full_name: Ritter, Alan
last_name: Ritter
- first_name: Lu
full_name: Wang, Lu
last_name: Wang
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://aclanthology.org/2025.naacl-long.122/
oa: '1'
page: 2421–2449
place: Albuquerque, New Mexico
project:
- _id: '118'
name: 'TRR 318: Project Area INF'
- _id: '126'
name: TRR 318 - Subproject C3
publication: 'Proceedings of the 2025 Conference of the Nations of the Americas Chapter
of the Association for Computational Linguistics: Human Language Technologies (Volume
1: Long Papers)'
publication_identifier:
isbn:
- 979-8-89176-189-6
publication_status: published
publisher: Association for Computational Linguistics
related_material:
link:
- relation: software
url: https://github.com/webis-de/naacl25-prompt-compositions
status: public
title: 'Adaptive Prompting: Ad-hoc Prompt Composition for Social Bias Detection'
type: conference
user_id: '84035'
year: '2025'
...
---
_id: '53073'
abstract:
- lang: eng
text: While shallow decision trees may be interpretable, larger ensemble models
like gradient-boosted trees, which often set the state of the art in machine learning
problems involving tabular data, still remain black box models. As a remedy, the
Shapley value (SV) is a well-known concept in explainable artificial intelligence
(XAI) research for quantifying additive feature attributions of predictions. The
model-specific TreeSHAP methodology solves the exponential complexity for retrieving
exact SVs from tree-based models. Expanding beyond individual feature attribution,
Shapley interactions reveal the impact of intricate feature interactions of any
order. In this work, we present TreeSHAP-IQ, an efficient method to compute any-order
additive Shapley interactions for predictions of tree-based models. TreeSHAP-IQ
is supported by a mathematical framework that exploits polynomial arithmetic to
compute the interaction scores in a single recursive traversal of the tree, akin
to Linear TreeSHAP. We apply TreeSHAP-IQ on state-of-the-art tree ensembles and
explore interactions on well-established benchmark datasets.
author:
- first_name: Maximilian
full_name: Muschalik, Maximilian
last_name: Muschalik
- first_name: Fabian
full_name: Fumagalli, Fabian
id: '93420'
last_name: Fumagalli
- first_name: Barbara
full_name: Hammer, Barbara
last_name: Hammer
- first_name: Eyke
full_name: Huellermeier, Eyke
id: '48129'
last_name: Huellermeier
citation:
ama: 'Muschalik M, Fumagalli F, Hammer B, Huellermeier E. Beyond TreeSHAP: Efficient
Computation of Any-Order Shapley Interactions for Tree Ensembles. In: Proceedings
of the AAAI Conference on Artificial Intelligence (AAAI). Vol 38. ; 2024:14388-14396.
doi:10.1609/aaai.v38i13.29352'
apa: 'Muschalik, M., Fumagalli, F., Hammer, B., & Huellermeier, E. (2024). Beyond
TreeSHAP: Efficient Computation of Any-Order Shapley Interactions for Tree Ensembles.
Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 38(13),
14388–14396. https://doi.org/10.1609/aaai.v38i13.29352'
bibtex: '@inproceedings{Muschalik_Fumagalli_Hammer_Huellermeier_2024, title={Beyond
TreeSHAP: Efficient Computation of Any-Order Shapley Interactions for Tree Ensembles},
volume={38}, DOI={10.1609/aaai.v38i13.29352},
number={13}, booktitle={Proceedings of the AAAI Conference on Artificial Intelligence
(AAAI)}, author={Muschalik, Maximilian and Fumagalli, Fabian and Hammer, Barbara
and Huellermeier, Eyke}, year={2024}, pages={14388–14396} }'
chicago: 'Muschalik, Maximilian, Fabian Fumagalli, Barbara Hammer, and Eyke Huellermeier.
“Beyond TreeSHAP: Efficient Computation of Any-Order Shapley Interactions for
Tree Ensembles.” In Proceedings of the AAAI Conference on Artificial Intelligence
(AAAI), 38:14388–96, 2024. https://doi.org/10.1609/aaai.v38i13.29352.'
ieee: 'M. Muschalik, F. Fumagalli, B. Hammer, and E. Huellermeier, “Beyond TreeSHAP:
Efficient Computation of Any-Order Shapley Interactions for Tree Ensembles,” in
Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2024,
vol. 38, no. 13, pp. 14388–14396, doi: 10.1609/aaai.v38i13.29352.'
mla: 'Muschalik, Maximilian, et al. “Beyond TreeSHAP: Efficient Computation of Any-Order
Shapley Interactions for Tree Ensembles.” Proceedings of the AAAI Conference
on Artificial Intelligence (AAAI), vol. 38, no. 13, 2024, pp. 14388–96, doi:10.1609/aaai.v38i13.29352.'
short: 'M. Muschalik, F. Fumagalli, B. Hammer, E. Huellermeier, in: Proceedings
of the AAAI Conference on Artificial Intelligence (AAAI), 2024, pp. 14388–14396.'
date_created: 2024-03-27T14:50:04Z
date_updated: 2025-09-11T16:20:11Z
department:
- _id: '660'
doi: 10.1609/aaai.v38i13.29352
intvolume: ' 38'
issue: '13'
keyword:
- Explainable Artificial Intelligence
language:
- iso: eng
page: 14388-14396
project:
- _id: '126'
name: 'TRR 318 - C3: TRR 318 - Subproject C3'
- _id: '109'
name: 'TRR 318: TRR 318 - Erklärbarkeit konstruieren'
- _id: '117'
name: 'TRR 318 - C: TRR 318 - Project Area C'
publication: Proceedings of the AAAI Conference on Artificial Intelligence (AAAI)
publication_identifier:
issn:
- 2374-3468
- 2159-5399
publication_status: published
status: public
title: 'Beyond TreeSHAP: Efficient Computation of Any-Order Shapley Interactions for
Tree Ensembles'
type: conference
user_id: '93420'
volume: 38
year: '2024'
...
---
_id: '55311'
abstract:
- lang: eng
text: Addressing the limitations of individual attribution scores via the Shapley
value (SV), the field of explainable AI (XAI) has recently explored intricate
interactions of features or data points. In particular, extensions of the SV,
such as the Shapley Interaction Index (SII), have been proposed as a measure to
still benefit from the axiomatic basis of the SV. However, similar to the SV,
their exact computation remains computationally prohibitive. Hence, we propose
with SVARM-IQ a sampling-based approach to efficiently approximate Shapley-based
interaction indices of any order. SVARM-IQ can be applied to a broad class of
interaction indices, including the SII, by leveraging a novel stratified representation.
We provide non-asymptotic theoretical guarantees on its approximation quality
and empirically demonstrate that SVARM-IQ achieves state-of-the-art estimation
results in practical XAI scenarios on different model classes and application
domains.
author:
- first_name: Patrick
full_name: Kolpaczki, Patrick
last_name: Kolpaczki
- first_name: Maximilian
full_name: Muschalik, Maximilian
last_name: Muschalik
- first_name: Fabian
full_name: Fumagalli, Fabian
id: '93420'
last_name: Fumagalli
- first_name: Barbara
full_name: Hammer, Barbara
last_name: Hammer
- first_name: Eyke
full_name: Huellermeier, Eyke
id: '48129'
last_name: Huellermeier
citation:
ama: 'Kolpaczki P, Muschalik M, Fumagalli F, Hammer B, Huellermeier E. SVARM-IQ:
Efficient Approximation of Any-order Shapley Interactions through Stratification.
In: Proceedings of The 27th International Conference on Artificial Intelligence
and Statistics (AISTATS). Vol 238. Proceedings of Machine Learning Research.
PMLR; 2024:3520–3528.'
apa: 'Kolpaczki, P., Muschalik, M., Fumagalli, F., Hammer, B., & Huellermeier,
E. (2024). SVARM-IQ: Efficient Approximation of Any-order Shapley Interactions
through Stratification. Proceedings of The 27th International Conference on
Artificial Intelligence and Statistics (AISTATS), 238, 3520–3528.'
bibtex: '@inproceedings{Kolpaczki_Muschalik_Fumagalli_Hammer_Huellermeier_2024,
series={Proceedings of Machine Learning Research}, title={SVARM-IQ: Efficient
Approximation of Any-order Shapley Interactions through Stratification}, volume={238},
booktitle={Proceedings of The 27th International Conference on Artificial Intelligence
and Statistics (AISTATS)}, publisher={PMLR}, author={Kolpaczki, Patrick and Muschalik,
Maximilian and Fumagalli, Fabian and Hammer, Barbara and Huellermeier, Eyke},
year={2024}, pages={3520–3528}, collection={Proceedings of Machine Learning Research}
}'
chicago: 'Kolpaczki, Patrick, Maximilian Muschalik, Fabian Fumagalli, Barbara Hammer,
and Eyke Huellermeier. “SVARM-IQ: Efficient Approximation of Any-Order Shapley
Interactions through Stratification.” In Proceedings of The 27th International
Conference on Artificial Intelligence and Statistics (AISTATS), 238:3520–3528.
Proceedings of Machine Learning Research. PMLR, 2024.'
ieee: 'P. Kolpaczki, M. Muschalik, F. Fumagalli, B. Hammer, and E. Huellermeier,
“SVARM-IQ: Efficient Approximation of Any-order Shapley Interactions through Stratification,”
in Proceedings of The 27th International Conference on Artificial Intelligence
and Statistics (AISTATS), 2024, vol. 238, pp. 3520–3528.'
mla: 'Kolpaczki, Patrick, et al. “SVARM-IQ: Efficient Approximation of Any-Order
Shapley Interactions through Stratification.” Proceedings of The 27th International
Conference on Artificial Intelligence and Statistics (AISTATS), vol. 238,
PMLR, 2024, pp. 3520–3528.'
short: 'P. Kolpaczki, M. Muschalik, F. Fumagalli, B. Hammer, E. Huellermeier, in:
Proceedings of The 27th International Conference on Artificial Intelligence and
Statistics (AISTATS), PMLR, 2024, pp. 3520–3528.'
date_created: 2024-07-18T09:39:14Z
date_updated: 2025-09-11T16:22:30Z
department:
- _id: '660'
intvolume: ' 238'
language:
- iso: eng
page: 3520–3528
project:
- _id: '109'
name: 'TRR 318: TRR 318 - Erklärbarkeit konstruieren'
- _id: '117'
name: 'TRR 318 - C: TRR 318 - Project Area C'
- _id: '126'
name: 'TRR 318 - C3: TRR 318 - Subproject C3'
publication: Proceedings of The 27th International Conference on Artificial Intelligence
and Statistics (AISTATS)
publisher: PMLR
series_title: Proceedings of Machine Learning Research
status: public
title: 'SVARM-IQ: Efficient Approximation of Any-order Shapley Interactions through
Stratification'
type: conference
user_id: '93420'
volume: 238
year: '2024'
...
---
_id: '61228'
author:
- first_name: Maximilian
full_name: Muschalik, Maximilian
last_name: Muschalik
- first_name: Hubert
full_name: Baniecki, Hubert
last_name: Baniecki
- first_name: Fabian
full_name: Fumagalli, Fabian
id: '93420'
last_name: Fumagalli
- first_name: Patrick
full_name: Kolpaczki, Patrick
last_name: Kolpaczki
- first_name: Barbara
full_name: Hammer, Barbara
last_name: Hammer
- first_name: Eyke
full_name: Huellermeier, Eyke
id: '48129'
last_name: Huellermeier
citation:
ama: 'Muschalik M, Baniecki H, Fumagalli F, Kolpaczki P, Hammer B, Huellermeier
E. shapiq: Shapley interactions for machine learning. In: Advances in Neural
Information Processing Systems (NeurIPS). Vol 37. ; 2024:130324–130357.'
apa: 'Muschalik, M., Baniecki, H., Fumagalli, F., Kolpaczki, P., Hammer, B., &
Huellermeier, E. (2024). shapiq: Shapley interactions for machine learning. Advances
in Neural Information Processing Systems (NeurIPS), 37, 130324–130357.'
bibtex: '@inproceedings{Muschalik_Baniecki_Fumagalli_Kolpaczki_Hammer_Huellermeier_2024,
title={shapiq: Shapley interactions for machine learning}, volume={37}, booktitle={Advances
in Neural Information Processing Systems (NeurIPS)}, author={Muschalik, Maximilian
and Baniecki, Hubert and Fumagalli, Fabian and Kolpaczki, Patrick and Hammer,
Barbara and Huellermeier, Eyke}, year={2024}, pages={130324–130357} }'
chicago: 'Muschalik, Maximilian, Hubert Baniecki, Fabian Fumagalli, Patrick Kolpaczki,
Barbara Hammer, and Eyke Huellermeier. “Shapiq: Shapley Interactions for Machine
Learning.” In Advances in Neural Information Processing Systems (NeurIPS),
37:130324–130357, 2024.'
ieee: 'M. Muschalik, H. Baniecki, F. Fumagalli, P. Kolpaczki, B. Hammer, and E.
Huellermeier, “shapiq: Shapley interactions for machine learning,” in Advances
in Neural Information Processing Systems (NeurIPS), 2024, vol. 37, pp. 130324–130357.'
mla: 'Muschalik, Maximilian, et al. “Shapiq: Shapley Interactions for Machine Learning.”
Advances in Neural Information Processing Systems (NeurIPS), vol. 37, 2024,
pp. 130324–130357.'
short: 'M. Muschalik, H. Baniecki, F. Fumagalli, P. Kolpaczki, B. Hammer, E. Huellermeier,
in: Advances in Neural Information Processing Systems (NeurIPS), 2024, pp. 130324–130357.'
date_created: 2025-09-11T15:39:01Z
date_updated: 2025-09-11T16:17:35Z
department:
- _id: '660'
intvolume: ' 37'
language:
- iso: eng
page: 130324–130357
project:
- _id: '117'
name: TRR 318 - Project Area C
- _id: '126'
name: TRR 318 - Subproject C3
- _id: '109'
name: 'TRR 318: Erklärbarkeit konstruieren'
publication: Advances in Neural Information Processing Systems (NeurIPS)
status: public
title: 'shapiq: Shapley interactions for machine learning'
type: conference
user_id: '93420'
volume: 37
year: '2024'
...
---
_id: '48778'
author:
- first_name: Maximilian
full_name: Muschalik, Maximilian
last_name: Muschalik
- first_name: Fabian
full_name: Fumagalli, Fabian
id: '93420'
last_name: Fumagalli
- first_name: Rohit
full_name: Jagtani, Rohit
last_name: Jagtani
- first_name: Barbara
full_name: Hammer, Barbara
last_name: Hammer
- first_name: Eyke
full_name: Huellermeier, Eyke
id: '48129'
last_name: Huellermeier
citation:
ama: 'Muschalik M, Fumagalli F, Jagtani R, Hammer B, Huellermeier E. iPDP: On Partial
Dependence Plots in Dynamic Modeling Scenarios. In: Proceedings of the World
Conference on Explainable Artificial Intelligence (XAI). ; 2023. doi:10.1007/978-3-031-44064-9_11'
apa: 'Muschalik, M., Fumagalli, F., Jagtani, R., Hammer, B., & Huellermeier,
E. (2023). iPDP: On Partial Dependence Plots in Dynamic Modeling Scenarios. Proceedings
of the World Conference on Explainable Artificial Intelligence (XAI). https://doi.org/10.1007/978-3-031-44064-9_11'
bibtex: '@inproceedings{Muschalik_Fumagalli_Jagtani_Hammer_Huellermeier_2023, title={iPDP:
On Partial Dependence Plots in Dynamic Modeling Scenarios}, DOI={10.1007/978-3-031-44064-9_11},
booktitle={Proceedings of the World Conference on Explainable Artificial Intelligence
(xAI)}, author={Muschalik, Maximilian and Fumagalli, Fabian and Jagtani, Rohit
and Hammer, Barbara and Huellermeier, Eyke}, year={2023} }'
chicago: 'Muschalik, Maximilian, Fabian Fumagalli, Rohit Jagtani, Barbara Hammer,
and Eyke Huellermeier. “IPDP: On Partial Dependence Plots in Dynamic Modeling
Scenarios.” In Proceedings of the World Conference on Explainable Artificial
Intelligence (XAI), 2023. https://doi.org/10.1007/978-3-031-44064-9_11.'
ieee: 'M. Muschalik, F. Fumagalli, R. Jagtani, B. Hammer, and E. Huellermeier, “iPDP:
On Partial Dependence Plots in Dynamic Modeling Scenarios,” 2023, doi: 10.1007/978-3-031-44064-9_11.'
mla: 'Muschalik, Maximilian, et al. “IPDP: On Partial Dependence Plots in Dynamic
Modeling Scenarios.” Proceedings of the World Conference on Explainable Artificial
Intelligence (XAI), 2023, doi:10.1007/978-3-031-44064-9_11.'
short: 'M. Muschalik, F. Fumagalli, R. Jagtani, B. Hammer, E. Huellermeier, in:
Proceedings of the World Conference on Explainable Artificial Intelligence (XAI),
2023.'
date_created: 2023-11-10T14:17:17Z
date_updated: 2025-09-11T16:14:34Z
department:
- _id: '660'
doi: 10.1007/978-3-031-44064-9_11
language:
- iso: eng
project:
- _id: '126'
name: 'TRR 318 - C3: TRR 318 - Subproject C3'
- _id: '109'
name: 'TRR 318: TRR 318 - Erklärbarkeit konstruieren'
- _id: '117'
name: 'TRR 318 - C: TRR 318 - Project Area C'
publication: Proceedings of the World Conference on Explainable Artificial Intelligence
(xAI)
publication_identifier:
eisbn:
- 1865-0937
eissn:
- '9783031440649'
isbn:
- '9783031440632'
issn:
- 1865-0929
publication_status: published
status: public
title: 'iPDP: On Partial Dependence Plots in Dynamic Modeling Scenarios'
type: conference
user_id: '93420'
year: '2023'
...
---
_id: '48776'
author:
- first_name: Maximilian
full_name: Muschalik, Maximilian
last_name: Muschalik
- first_name: Fabian
full_name: Fumagalli, Fabian
id: '93420'
last_name: Fumagalli
- first_name: Barbara
full_name: Hammer, Barbara
last_name: Hammer
- first_name: Eyke
full_name: Huellermeier, Eyke
id: '48129'
last_name: Huellermeier
citation:
ama: 'Muschalik M, Fumagalli F, Hammer B, Huellermeier E. iSAGE: An Incremental
Version of SAGE for Online Explanation on Data Streams. In: Machine Learning
and Knowledge Discovery in Databases: Research Track - European Conference (ECML
PKDD). Springer Nature Switzerland; 2023. doi:10.1007/978-3-031-43418-1_26'
apa: 'Muschalik, M., Fumagalli, F., Hammer, B., & Huellermeier, E. (2023). iSAGE:
An Incremental Version of SAGE for Online Explanation on Data Streams. In Machine
Learning and Knowledge Discovery in Databases: Research Track - European Conference
(ECML PKDD). Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-43418-1_26'
bibtex: '@inbook{Muschalik_Fumagalli_Hammer_Huellermeier_2023, title={iSAGE: An
Incremental Version of SAGE for Online Explanation on Data Streams}, DOI={10.1007/978-3-031-43418-1_26},
booktitle={Machine Learning and Knowledge Discovery in Databases: Research Track
- European Conference (ECML PKDD)}, publisher={Springer Nature Switzerland}, author={Muschalik,
Maximilian and Fumagalli, Fabian and Hammer, Barbara and Huellermeier, Eyke},
year={2023} }'
chicago: 'Muschalik, Maximilian, Fabian Fumagalli, Barbara Hammer, and Eyke Huellermeier.
“ISAGE: An Incremental Version of SAGE for Online Explanation on Data Streams.”
In Machine Learning and Knowledge Discovery in Databases: Research Track -
European Conference (ECML PKDD). Springer Nature Switzerland, 2023. https://doi.org/10.1007/978-3-031-43418-1_26.'
ieee: 'M. Muschalik, F. Fumagalli, B. Hammer, and E. Huellermeier, “iSAGE: An Incremental
Version of SAGE for Online Explanation on Data Streams,” in Machine Learning
and Knowledge Discovery in Databases: Research Track - European Conference (ECML
PKDD), Springer Nature Switzerland, 2023.'
mla: 'Muschalik, Maximilian, et al. “ISAGE: An Incremental Version of SAGE for Online
Explanation on Data Streams.” Machine Learning and Knowledge Discovery in Databases:
Research Track - European Conference (ECML PKDD), Springer Nature Switzerland,
2023, doi:10.1007/978-3-031-43418-1_26.'
short: 'M. Muschalik, F. Fumagalli, B. Hammer, E. Huellermeier, in: Machine Learning
and Knowledge Discovery in Databases: Research Track - European Conference (ECML
PKDD), Springer Nature Switzerland, 2023.'
date_created: 2023-11-10T14:11:20Z
date_updated: 2025-09-11T16:27:26Z
department:
- _id: '660'
doi: 10.1007/978-3-031-43418-1_26
language:
- iso: eng
project:
- _id: '126'
name: 'TRR 318 - C3: TRR 318 - Subproject C3'
- _id: '117'
name: 'TRR 318 - C: TRR 318 - Project Area C'
- _id: '109'
name: 'TRR 318: TRR 318 - Erklärbarkeit konstruieren'
publication: 'Machine Learning and Knowledge Discovery in Databases: Research Track
- European Conference (ECML PKDD)'
publication_identifier:
eisbn:
- '9783031434181'
eissn:
- 1611-3349
isbn:
- '9783031434174'
issn:
- 0302-9743
publication_status: published
publisher: Springer Nature Switzerland
status: public
title: 'iSAGE: An Incremental Version of SAGE for Online Explanation on Data Streams'
type: book_chapter
user_id: '93420'
year: '2023'
...
---
_id: '48775'
author:
- first_name: Fabian
full_name: Fumagalli, Fabian
id: '93420'
last_name: Fumagalli
- first_name: Maximilian
full_name: Muschalik, Maximilian
last_name: Muschalik
- first_name: Eyke
full_name: Hüllermeier, Eyke
id: '48129'
last_name: Hüllermeier
- first_name: Barbara
full_name: Hammer, Barbara
last_name: Hammer
citation:
ama: 'Fumagalli F, Muschalik M, Hüllermeier E, Hammer B. On Feature Removal for
Explainability in Dynamic Environments. In: Proceedings of the European Symposium
on Artificial Neural Networks (ESANN). ; 2023. doi:10.14428/ESANN/2023.ES2023-148'
apa: Fumagalli, F., Muschalik, M., Hüllermeier, E., & Hammer, B. (2023). On
Feature Removal for Explainability in Dynamic Environments. Proceedings of
the European Symposium on Artificial Neural Networks (ESANN). ESANN 2023 -
European Symposium on Artificial Neural Networks, Computational Intelligence and
Machine Learning, Bruges (Belgium) and online. https://doi.org/10.14428/ESANN/2023.ES2023-148
bibtex: '@inproceedings{Fumagalli_Muschalik_Hüllermeier_Hammer_2023, title={On Feature
Removal for Explainability in Dynamic Environments}, DOI={10.14428/ESANN/2023.ES2023-148},
booktitle={Proceedings of the European Symposium on Artificial Neural Networks
(ESANN)}, author={Fumagalli, Fabian and Muschalik, Maximilian and Hüllermeier,
Eyke and Hammer, Barbara}, year={2023} }'
chicago: Fumagalli, Fabian, Maximilian Muschalik, Eyke Hüllermeier, and Barbara
Hammer. “On Feature Removal for Explainability in Dynamic Environments.” In Proceedings
of the European Symposium on Artificial Neural Networks (ESANN), 2023. https://doi.org/10.14428/ESANN/2023.ES2023-148.
ieee: 'F. Fumagalli, M. Muschalik, E. Hüllermeier, and B. Hammer, “On Feature Removal
for Explainability in Dynamic Environments,” presented at the ESANN 2023 - European
Symposium on Artificial Neural Networks, Computational Intelligence and Machine
Learning, Bruges (Belgium) and online, 2023, doi: 10.14428/ESANN/2023.ES2023-148.'
mla: Fumagalli, Fabian, et al. “On Feature Removal for Explainability in Dynamic
Environments.” Proceedings of the European Symposium on Artificial Neural Networks
(ESANN), 2023, doi:10.14428/ESANN/2023.ES2023-148.
short: 'F. Fumagalli, M. Muschalik, E. Hüllermeier, B. Hammer, in: Proceedings of
the European Symposium on Artificial Neural Networks (ESANN), 2023.'
conference:
location: Bruges (Belgium) and online
name: ESANN 2023 - European Symposium on Artificial Neural Networks, Computational
Intelligence and Machine Learning
date_created: 2023-11-10T14:00:08Z
date_updated: 2025-09-11T16:26:21Z
department:
- _id: '660'
doi: 10.14428/ESANN/2023.ES2023-148
language:
- iso: eng
project:
- _id: '126'
name: 'TRR 318 - C3: TRR 318 - Subproject C3'
- _id: '117'
name: 'TRR 318 - C: TRR 318 - Project Area C'
- _id: '109'
name: 'TRR 318: TRR 318 - Erklärbarkeit konstruieren'
publication: Proceedings of the European Symposium on Artificial Neural Networks (ESANN)
publication_identifier:
unknown:
- ' 978-2-87587-088-9'
publication_status: published
status: public
title: On Feature Removal for Explainability in Dynamic Environments
type: conference
user_id: '93420'
year: '2023'
...
---
_id: '52230'
author:
- first_name: Fabian
full_name: Fumagalli, Fabian
id: '93420'
last_name: Fumagalli
- first_name: Maximilian
full_name: Muschalik, Maximilian
last_name: Muschalik
- first_name: Patrick
full_name: Kolpaczki, Patrick
last_name: Kolpaczki
- first_name: Eyke
full_name: Hüllermeier, Eyke
id: '48129'
last_name: Hüllermeier
- first_name: Barbara
full_name: Hammer, Barbara
last_name: Hammer
citation:
ama: 'Fumagalli F, Muschalik M, Kolpaczki P, Hüllermeier E, Hammer B. SHAP-IQ: Unified
Approximation of any-order Shapley Interactions. In: Advances in Neural Information
Processing Systems (NeurIPS). Vol 36. ; 2023:11515--11551.'
apa: 'Fumagalli, F., Muschalik, M., Kolpaczki, P., Hüllermeier, E., & Hammer,
B. (2023). SHAP-IQ: Unified Approximation of any-order Shapley Interactions. Advances
in Neural Information Processing Systems (NeurIPS), 36, 11515--11551.'
bibtex: '@inproceedings{Fumagalli_Muschalik_Kolpaczki_Hüllermeier_Hammer_2023, title={SHAP-IQ:
Unified Approximation of any-order Shapley Interactions}, volume={36}, booktitle={Advances
in Neural Information Processing Systems (NeurIPS)}, author={Fumagalli, Fabian
and Muschalik, Maximilian and Kolpaczki, Patrick and Hüllermeier, Eyke and Hammer,
Barbara}, year={2023}, pages={11515--11551} }'
chicago: 'Fumagalli, Fabian, Maximilian Muschalik, Patrick Kolpaczki, Eyke Hüllermeier,
and Barbara Hammer. “SHAP-IQ: Unified Approximation of Any-Order Shapley Interactions.”
In Advances in Neural Information Processing Systems (NeurIPS), 36:11515--11551,
2023.'
ieee: 'F. Fumagalli, M. Muschalik, P. Kolpaczki, E. Hüllermeier, and B. Hammer,
“SHAP-IQ: Unified Approximation of any-order Shapley Interactions,” in Advances
in Neural Information Processing Systems (NeurIPS), 2023, vol. 36, pp. 11515--11551.'
mla: 'Fumagalli, Fabian, et al. “SHAP-IQ: Unified Approximation of Any-Order Shapley
Interactions.” Advances in Neural Information Processing Systems (NeurIPS),
vol. 36, 2023, pp. 11515--11551.'
short: 'F. Fumagalli, M. Muschalik, P. Kolpaczki, E. Hüllermeier, B. Hammer, in:
Advances in Neural Information Processing Systems (NeurIPS), 2023, pp. 11515--11551.'
date_created: 2024-03-01T14:15:31Z
date_updated: 2025-09-11T16:18:16Z
department:
- _id: '660'
intvolume: ' 36'
language:
- iso: eng
page: 11515--11551
project:
- _id: '126'
name: 'TRR 318 - C3: TRR 318 - Subproject C3'
- _id: '109'
name: 'TRR 318: TRR 318 - Erklärbarkeit konstruieren'
- _id: '117'
name: 'TRR 318 - C: TRR 318 - Project Area C'
publication: Advances in Neural Information Processing Systems (NeurIPS)
status: public
title: 'SHAP-IQ: Unified Approximation of any-order Shapley Interactions'
type: conference
user_id: '93420'
volume: 36
year: '2023'
...
---
_id: '48780'
abstract:
- lang: eng
text: Explainable Artificial Intelligence (XAI) has mainly focused on static learning
tasks so far. In this paper, we consider XAI in the context of online learning
in dynamic environments, such as learning from real-time data streams, where models
are learned incrementally and continuously adapted over the course of time. More
specifically, we motivate the problem of explaining model change, i.e. explaining
the difference between models before and after adaptation, instead of the models
themselves. In this regard, we provide the first efficient model-agnostic approach
to dynamically detecting, quantifying, and explaining significant model changes.
Our approach is based on an adaptation of the well-known Permutation Feature Importance
(PFI) measure. It includes two hyperparameters that control the sensitivity and
directly influence explanation frequency, so that a human user can adjust the
method to individual requirements and application needs. We assess and validate
our method’s efficacy on illustrative synthetic data streams with three popular
model classes.
author:
- first_name: Maximilian
full_name: Muschalik, Maximilian
last_name: Muschalik
- first_name: Fabian
full_name: Fumagalli, Fabian
id: '93420'
last_name: Fumagalli
- first_name: Barbara
full_name: Hammer, Barbara
last_name: Hammer
- first_name: Eyke
full_name: Huellermeier, Eyke
id: '48129'
last_name: Huellermeier
citation:
ama: Muschalik M, Fumagalli F, Hammer B, Huellermeier E. Agnostic Explanation of
Model Change based on Feature Importance. KI - Künstliche Intelligenz.
2022;36(3-4):211-224. doi:10.1007/s13218-022-00766-6
apa: Muschalik, M., Fumagalli, F., Hammer, B., & Huellermeier, E. (2022). Agnostic
Explanation of Model Change based on Feature Importance. KI - Künstliche Intelligenz,
36(3–4), 211–224. https://doi.org/10.1007/s13218-022-00766-6
bibtex: '@article{Muschalik_Fumagalli_Hammer_Huellermeier_2022, title={Agnostic
Explanation of Model Change based on Feature Importance}, volume={36}, DOI={10.1007/s13218-022-00766-6},
number={3–4}, journal={KI - Künstliche Intelligenz}, publisher={Springer Science
and Business Media LLC}, author={Muschalik, Maximilian and Fumagalli, Fabian and
Hammer, Barbara and Huellermeier, Eyke}, year={2022}, pages={211–224} }'
chicago: 'Muschalik, Maximilian, Fabian Fumagalli, Barbara Hammer, and Eyke Huellermeier.
“Agnostic Explanation of Model Change Based on Feature Importance.” KI - Künstliche
Intelligenz 36, no. 3–4 (2022): 211–24. https://doi.org/10.1007/s13218-022-00766-6.'
ieee: 'M. Muschalik, F. Fumagalli, B. Hammer, and E. Huellermeier, “Agnostic Explanation
of Model Change based on Feature Importance,” KI - Künstliche Intelligenz,
vol. 36, no. 3–4, pp. 211–224, 2022, doi: 10.1007/s13218-022-00766-6.'
mla: Muschalik, Maximilian, et al. “Agnostic Explanation of Model Change Based on
Feature Importance.” KI - Künstliche Intelligenz, vol. 36, no. 3–4, Springer
Science and Business Media LLC, 2022, pp. 211–24, doi:10.1007/s13218-022-00766-6.
short: M. Muschalik, F. Fumagalli, B. Hammer, E. Huellermeier, KI - Künstliche Intelligenz
36 (2022) 211–224.
date_created: 2023-11-10T14:21:06Z
date_updated: 2025-01-16T16:19:35Z
department:
- _id: '660'
doi: 10.1007/s13218-022-00766-6
intvolume: ' 36'
issue: 3-4
keyword:
- Artificial Intelligence
language:
- iso: eng
page: 211-224
project:
- _id: '126'
name: 'TRR 318 - C3: TRR 318 - Subproject C3'
- _id: '117'
name: 'TRR 318 - C: TRR 318 - Project Area C'
- _id: '109'
grant_number: '438445824'
name: 'TRR 318: TRR 318 - Erklärbarkeit konstruieren'
publication: KI - Künstliche Intelligenz
publication_identifier:
issn:
- 0933-1875
- 1610-1987
publication_status: published
publisher: Springer Science and Business Media LLC
status: public
title: Agnostic Explanation of Model Change based on Feature Importance
type: journal_article
user_id: '93420'
volume: 36
year: '2022'
...