Lune

ICML2023Top-tier venue

A Model-free Closeness-of-influence Test for Features in Supervised Learning

Mohammad Mehrabi, Ryan A. Rossi

Abstract

Understanding the effect of a feature vector x∈Rdx \in \mathbb{R}^d on the response value (label) y∈Ry \in \mathbb{R} is the cornerstone of many statistical learning problems. Ideally, it is desired to understand how a set of collected features combine together and influence the response value, but this problem is notoriously difficult, due to the high-dimensionality of data and limited number of labeled data points, among many others. In this work, we take a new perspective on this problem, and we study the question of assessing the difference of influence that the two given features have on the response value. We first propose a notion of closeness for the influence of features, and show that our definition recovers the familiar notion of the magnitude of coefficients in the parametric model. We then propose a novel method to test for the closeness of influence in general model-free supervised learning problems. Our proposed test can be used with finite number of samples with control on type I error rate, no matter the ground truth conditional law L(Y∣X)\mathcal{L}(Y |X). We analyze the power of our test for two general learning problems i) linear regression, and ii) binary classification under mixture of Gaussian models, and show that under the proper choice of score function, an internal component of our test, with sufficient number of samples will achieve full statistical power. We evaluate our findings through extensive numerical simulations, specifically we adopt the datamodel framework (Ilyas, et al., 2022) for CIFAR-10 dataset to identify pairs of training samples with different influence on the trained model via optional black box training mechanisms.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

1 top-tier paper has cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers1

  1. Are Two Datasets Close Enough With Statistical Significance? A Kernel Distributional Closeness Testing ApproachZhijian Zhou, Liuhua Peng, Xunye Tian, Mingming Gong et al.ICML 2026 · 1 citation

Related papers

  1. On Second-Order Group Influence Functions for Black-Box PredictionsSamyadeep Basu, Xuchen You, Soheil FeiziICML 2020 · 28 citations
  2. Understanding Influence Functions and Datamodels via Harmonic AnalysisNikunj Saunshi, Arushi Gupta, Mark Braverman, Sanjeev AroraICLR 2023 · 1 citation
  3. Evaluating Data Influence in Meta LearningChenyang Ren, Huanyi Xie, Shu Yang, Meng Ding et al.ICLR 2026 · 5 citations
  4. Enhancing Prediction Performance through Influence MeasureShuguang Yu, Wenqian Xu, Xinyi Zhou, Xuechun Wang et al.ICLR 2025
  5. "What Data Benefits My Classifier?" Enhancing Model Performance and Interpretability through Influence-Based Data SelectionAnshuman Chhabra, Peizhao Li, Prasant Mohapatra, Hongfu LiuICLR 2024 · 32 citations
Dusk over the sea between two cliffs drawn in fine vertical lines