Extending Fair Null-Space Projections for Continuous Attributes to Kernel Methods

Felix Störck, Fabian Hinder, Barbara Hammer
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:115812-115825, 2026.

Abstract

With the on-going integration of machine learning systems into the everyday social life of millions the notion of fairness becomes an ever increasing priority in their development. Fairness notions commonly rely on protected attributes to assess potential biases. Here, the majority of literature focuses on discrete setups regarding both target and protected attributes. The literature on continuous attributes especially in conjunction with regression - we refer to this as continuous fairness - is scarce. A common strategy is iterative null-space projection which as of now has only been explored for linear models or embeddings such as obtained by a non-linear encoder. We improve on this by extending this to kernel induced feature spaces by means of the “empirical feature space”. We theoretically derive this as a direct transformation of the kernel matrix yielding a model and fairness-score agnostic method applicable to continuous protected attributes. We demonstrate that our novel approach in conjunction with Support Vector Regression (SVR) provides competitive or improved performance across multiple datasets in comparison to other contemporary methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-storck26a, title = {Extending Fair Null-Space Projections for Continuous Attributes to Kernel Methods}, author = {St\"{o}rck, Felix and Hinder, Fabian and Hammer, Barbara}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {115812--115825}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/storck26a/storck26a.pdf}, url = {https://proceedings.mlr.press/v306/storck26a.html}, abstract = {With the on-going integration of machine learning systems into the everyday social life of millions the notion of fairness becomes an ever increasing priority in their development. Fairness notions commonly rely on protected attributes to assess potential biases. Here, the majority of literature focuses on discrete setups regarding both target and protected attributes. The literature on continuous attributes especially in conjunction with regression - we refer to this as continuous fairness - is scarce. A common strategy is iterative null-space projection which as of now has only been explored for linear models or embeddings such as obtained by a non-linear encoder. We improve on this by extending this to kernel induced feature spaces by means of the “empirical feature space”. We theoretically derive this as a direct transformation of the kernel matrix yielding a model and fairness-score agnostic method applicable to continuous protected attributes. We demonstrate that our novel approach in conjunction with Support Vector Regression (SVR) provides competitive or improved performance across multiple datasets in comparison to other contemporary methods.} }
Endnote
%0 Conference Paper %T Extending Fair Null-Space Projections for Continuous Attributes to Kernel Methods %A Felix Störck %A Fabian Hinder %A Barbara Hammer %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-storck26a %I PMLR %P 115812--115825 %U https://proceedings.mlr.press/v306/storck26a.html %V 306 %X With the on-going integration of machine learning systems into the everyday social life of millions the notion of fairness becomes an ever increasing priority in their development. Fairness notions commonly rely on protected attributes to assess potential biases. Here, the majority of literature focuses on discrete setups regarding both target and protected attributes. The literature on continuous attributes especially in conjunction with regression - we refer to this as continuous fairness - is scarce. A common strategy is iterative null-space projection which as of now has only been explored for linear models or embeddings such as obtained by a non-linear encoder. We improve on this by extending this to kernel induced feature spaces by means of the “empirical feature space”. We theoretically derive this as a direct transformation of the kernel matrix yielding a model and fairness-score agnostic method applicable to continuous protected attributes. We demonstrate that our novel approach in conjunction with Support Vector Regression (SVR) provides competitive or improved performance across multiple datasets in comparison to other contemporary methods.
APA
Störck, F., Hinder, F. & Hammer, B.. (2026). Extending Fair Null-Space Projections for Continuous Attributes to Kernel Methods. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:115812-115825 Available from https://proceedings.mlr.press/v306/storck26a.html.

Related Material