Generalized Functional ANOVA: A Complete Theoretical Framework

Abstract

The functional ANOVA provides a fundamental representation of square-integrable multivariate functions into main effects and higher-order interactions. For independent inputs, the components belong to mutually orthogonal Hilbert subspaces and admit an explicit representation. For dependent inputs, however, the components are only hierarchically orthogonal: although existence and uniqueness results are available, the Hilbert subspaces underlying the generalized decomposition have remained implicit. We resolve this representation problem for continuous inputs supported on a bounded hyperrectangle whose joint density is bounded above and away from zero. We introduce a distribution-adapted family of functions and prove that it forms a Riesz basis of the L² space, thereby guaranteeing a unique, stable, and unconditionally convergent representation. We then show that, for every coalition of variables, the corresponding block of this basis exactly characterizes the Hilbert subspace containing the functional ANOVA component. Our construction recovers the classical orthogonal decomposition under input independence. As a direct consequence, computing the generalized functional ANOVA reduces to estimating coefficients in an explicit, distribution-adapted basis. Finally, as a \emph{proof of concept}, we introduce an elementary, fast and model-agnostic estimator based on our theoretical results. Experiments on synthetic and real-world datasets illustrate its connections with established tabular explanation methods and show that low order components often capture most of the signal in the model output.

Generalized property F for weakly group-theoretical modular fusion categories
Enyu Li, Yuze Ruan
arXiv4 Oct 2026PreprintarXiv licenceFull textAlgebra & number theoryarXiv
A complete classification for n-species Lotka-Volterra stably dissipative systems based on a generalized AHU algorithm
Yajia Du, Zhengyi Lu, Jinya Xu +1
arXiv1 Nov 2026PreprintCC BY 4.0Full textDynamical systemsarXiv
A Complete Equational Theory for Quantum Circuits with Generalized Control
William Schober, Scott Wesley
arXiv4 Nov 2026PreprintCC BY 4.0Full textAlgebra & number theoryarXiv
On Some Theoretical Aspects of Functional Data Analysis of Parametrized Curves in ℝ^p
Bernard Colin
arXiv23 Sep 2026PreprintCC BY-NC-ND 4.0Full textStatisticsarXiv
Complete mapping criteria for generalized Cesàro operators between Hardy spaces
Pengcheng Tang, Huayou Xie
arXiv17 Sep 2026PreprintCC BY-NC-SA 4.0Full textAnalysis & PDEsarXiv

Cite

BibTeX RIS

BibTeX

@misc{ferrere2026generalized,
  title = {{Generalized Functional ANOVA: A Complete Theoretical Framework}},
  author = {Baptiste Ferrere and Nicolas Bousquet and Fabrice Gamboa and Jean-Michel Loubes},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2605.18422},
  eprint = {2605.18422},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2605.18422},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Generalized Functional ANOVA: A Complete Theoretical Framework
AU  - Baptiste Ferrere
AU  - Nicolas Bousquet
AU  - Fabrice Gamboa
AU  - Jean-Michel Loubes
PY  - 2026
DA  - 2026/09/26
PB  - arXiv
DO  - 10.48550/arxiv.2605.18422
UR  - https://arxiv.org/abs/2605.18422
AB  - The functional ANOVA provides a fundamental representation of square-integrable multivariate functions into main effects and higher-order interactions. For independent inputs, the components belong to mutually orthogonal Hilbert subspaces and admit an explicit representation. For dependent inputs, however, the components are only hierarchically orthogonal: although existence and uniqueness results are available, the Hilbert subspaces underlying the generalized decomposition have remained implicit. We resolve this representation problem for continuous inputs supported on a bounded hyperrectangle whose joint density is bounded above and away from zero. We introduce a distribution-adapted family of functions and prove that it forms a Riesz basis of the L^2 space, thereby guaranteeing a unique, stable, and unconditionally convergent representation. We then show that, for every coalition of variables, the corresponding block of this basis exactly characterizes the Hilbert subspace containing the functional ANOVA component. Our construction recovers the classical orthogonal decomposition under input independence. As a direct consequence, computing the generalized functional ANOVA reduces to estimating coefficients in an explicit, distribution-adapted basis. Finally, as a \emph{proof of concept}, we introduce an elementary, fast and model-agnostic estimator based on our theoretical results. Experiments on synthetic and real-world datasets illustrate its connections with established tabular explanation methods and show that low order components often capture most of the signal in the model output.
ER  -

Abstract

Related

Cite