Searcharxiv⌕ Search

arXiv subjects

Chenghui Zheng

Publications and source records attributed to Chenghui Zheng.

4 recordsLinked to original sources
arXiv2026

Comparing Model-agnostic Feature Selection Methods through Relative Efficiency

Feature selection and importance estimation in a model-agnostic setting is an ongoing challenge of significant interest. Wrapper methods are commonly used because they are typically model-agnostic. In this paper, we develop a general comparison framework for model-agnostic feature selection methods based on relative efficiency, using \emph{relative variability} $σ/μ$ to account for different statistics having different means. In particular we focus on state-of-the-art feature selection methods, the Generalized Covariance Measure (GCM) and Leave-One-Covariate-Out (LOCO) estimation. In particular, we present a theoretical comparison under three model settings: linear models, non-linear additive models, and single index models that mimic a single-layer neural network. We complement this with simulations and real data examples for the above models and mis-specified models. Our theoretical results, along with empirical findings, demonstrate that GCM-related methods generally out-perform LOCO under suitable regularity conditions defined by a suitably defined correlation quantity which quantifies the asymptotic relative efficiency of these approaches. Our simulations and real data analysis include widely used machine learning methods such as neural networks and gradient boosting trees.

stat.ML↗
arXiv2026

MinShap: A Shapley-Based Framework for Feature Redundancy

Shapley values provide a flexible framework for attributing feature contributions to model predictions, but they are not naturally suited for feature selection: a feature may receive a positive attribution even when it is redundant given the remaining variables. In this paper, we introduce \textbf{MinShap}, a general framework for identifying \emph{important} or \emph{non-redundant} features through conditional importance functionals $VI_j^S$. Rather than averaging feature contributions across conditioning sets, MinShap aggregates them using the \emph{minimum}, thereby testing whether a feature remains relevant under every conditioning context. We show that, under a simple \emph{null monotonicity} condition, the minimum aggregation exactly characterizes feature redundancy and yields a principled feature selection criterion. This perspective provides a unified framework for statistical feature selection and representation-based interpretability while retaining the stability advantages of Shapley-style aggregation. We develop scalable algorithms with statistical guarantees, establish connections to multiple-testing procedures, and demonstrate through theory and experiments that MinShap produces more accurate and stable feature selection than existing model-agnostic approaches.

stat.ML↗
arXiv2021

A conjectural asymptotic formula for multiplicative chaos in number theory

We investigate a special sequence of random variables $A(N)$ defined by an exponential power series with independent standard complex Gaussians $(X(k))_{k \geq 1}$. Introduced by Hughes, Keating, and O'Connell in the study of random matrix theory, this sequence relates to Gaussian multiplicative chaos (in particular "holomorphic multiplicative chaos'' per Najnudel, Paquette, and Simm) and random multiplicative functions. Soundararajan and Zaman recently determined the order of $\mathbb{E}[|A(N)|]$. By constructing an algorithm to calculate $A(N)$ in $O(N^2 \log N)$ steps, we produce computational evidence that their result can likely be strengthened to an asymptotic result with a numerical estimate for the asymptotic constant. We also obtain similar conclusions when $A(N)$ is defined using standard real Gaussians or uniform $\pm 1$ random variables. However, our evidence suggests that the asymptotic constants do not possess a natural product structure.

math.NT↗