Search papers, labs, and topics across Lattice.
This paper addresses the challenge of computing exact Functional ANOVA (fANOVA) decompositions for models with categorical inputs, which was previously limited by the lack of closed-form expressions for dependent distributions. The authors derive a closed-form fANOVA decomposition for categorical inputs by connecting functional analysis with discrete Fourier analysis, removing the need for sampling-based approximations. This novel formulation is computationally efficient, handles arbitrary dependence structures, and provides a generalization of SHAP values for categorical features under dependence.
Forget approximate sampling: a closed-form solution now unlocks exact Functional ANOVA decomposition for categorical input models, even with complex feature dependencies.
Functional ANOVA offers a principled framework for interpretability by decomposing a model's prediction into main effects and higher-order interactions. For independent features, this decomposition is well-defined, strongly linked with SHAP values, and serves as a cornerstone of additive explainability. However, the lack of an explicit closed-form expression for general dependent distributions has forced practitioners to rely on costly sampling-based approximations. We completely resolve this limitation for categorical inputs. By bridging functional analysis with the extension of discrete Fourier analysis, we derive a closed-form decomposition without any assumption. Our formulation is computationally very efficient. It seamlessly recovers the classical independent case and extends to arbitrary dependence structures, including distributions with non-rectangular support. Furthermore, leveraging the intrinsic link between SHAP and ANOVA under independence, our framework yields a natural generalization of SHAP values for the general categorical setting.
