Statistical Significance of Feature Importance Rankings
- URL: http://arxiv.org/abs/2401.15800v4
- Date: Thu, 30 Jan 2025 23:25:22 GMT
- Title: Statistical Significance of Feature Importance Rankings
- Authors: Jeremy Goldwasser, Giles Hooker,
- Abstract summary: We devise techniques that ensure the most important features are correct with high-probability guarantees.
These assess the set of $K$ top-ranked features, as well as the order of its elements.
We then introduce two efficient sampling algorithms that identify the $K$ most important features, perhaps in order, with probability exceeding $1-alpha$.
- Score: 3.8642937395065124
- License:
- Abstract: Feature importance scores are ubiquitous tools for understanding the predictions of machine learning models. However, many popular attribution methods suffer from high instability due to random sampling. Leveraging novel ideas from hypothesis testing, we devise techniques that ensure the most important features are correct with high-probability guarantees. These assess the set of $K$ top-ranked features, as well as the order of its elements. Given a set of local or global importance scores, we demonstrate how to retrospectively verify the stability of the highest ranks. We then introduce two efficient sampling algorithms that identify the $K$ most important features, perhaps in order, with probability exceeding $1-\alpha$. The theoretical justification for these procedures is validated empirically on SHAP and LIME.
Related papers
- Probably Approximately Precision and Recall Learning [62.912015491907994]
Precision and Recall are foundational metrics in machine learning.
One-sided feedback--where only positive examples are observed during training--is inherent in many practical problems.
We introduce a PAC learning framework where each hypothesis is represented by a graph, with edges indicating positive interactions.
arXiv Detail & Related papers (2024-11-20T04:21:07Z) - The Certainty Ratio $C_ρ$: a novel metric for assessing the reliability of classifier predictions [0.0]
This paper introduces the Certainty Ratio ($C_rho$), a novel metric designed to quantify the contribution of confident (certain) versus uncertain predictions to any classification performance measure.
Experimental results across 21 datasets and multiple classifiers, including Decision Trees, Naive-Bayes, 3-Nearest Neighbors, and Random Forests, demonstrate that $C_rho$rho reveals critical insights that conventional metrics often overlook.
arXiv Detail & Related papers (2024-11-04T10:50:03Z) - A Probabilistic Perspective on Unlearning and Alignment for Large Language Models [48.96686419141881]
We introduce the first formal probabilistic evaluation framework for Large Language Models (LLMs)
Namely, we propose novel metrics with high probability guarantees concerning the output distribution of a model.
Our metrics are application-independent and allow practitioners to make more reliable estimates about model capabilities before deployment.
arXiv Detail & Related papers (2024-10-04T15:44:23Z) - Online non-parametric likelihood-ratio estimation by Pearson-divergence
functional minimization [55.98760097296213]
We introduce a new framework for online non-parametric LRE (OLRE) for the setting where pairs of iid observations $(x_t sim p, x'_t sim q)$ are observed over time.
We provide theoretical guarantees for the performance of the OLRE method along with empirical validation in synthetic experiments.
arXiv Detail & Related papers (2023-11-03T13:20:11Z) - Confident Feature Ranking [2.0564549686015594]
We present a framework for quantifying the uncertainty in global importance values.
We propose a novel method for the post-hoc interpretation of feature importance values.
arXiv Detail & Related papers (2023-07-28T07:23:01Z) - Generalized Differentiable RANSAC [95.95627475224231]
$nabla$-RANSAC is a differentiable RANSAC that allows learning the entire randomized robust estimation pipeline.
$nabla$-RANSAC is superior to the state-of-the-art in terms of accuracy while running at a similar speed to its less accurate alternatives.
arXiv Detail & Related papers (2022-12-26T15:13:13Z) - Don't Explain Noise: Robust Counterfactuals for Randomized Ensembles [50.81061839052459]
We formalize the generation of robust counterfactual explanations as a probabilistic problem.
We show the link between the robustness of ensemble models and the robustness of base learners.
Our method achieves high robustness with only a small increase in the distance from counterfactual explanations to their initial observations.
arXiv Detail & Related papers (2022-05-27T17:28:54Z) - On the Trustworthiness of Tree Ensemble Explainability Methods [0.9558392439655014]
Feature importance methods (e.g. gain and SHAP) are among the most popular explainability methods used to address this need.
For any explainability technique to be trustworthy and meaningful, it has to provide an explanation that is accurate and stable.
We evaluate the accuracy and stability of global feature importance methods through comprehensive experiments done on simulations and four real-world datasets.
arXiv Detail & Related papers (2021-09-30T20:56:37Z) - Learning to Rank Anomalies: Scalar Performance Criteria and Maximization
of Two-Sample Rank Statistics [0.0]
We propose a data-driven scoring function defined on the feature space which reflects the degree of abnormality of the observations.
This scoring function is learnt through a well-designed binary classification problem.
We illustrate our methodology with preliminary encouraging numerical experiments.
arXiv Detail & Related papers (2021-09-20T14:45:56Z) - Causal Feature Selection for Algorithmic Fairness [61.767399505764736]
We consider fairness in the integration component of data management.
We propose an approach to identify a sub-collection of features that ensure the fairness of the dataset.
arXiv Detail & Related papers (2020-06-10T20:20:10Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.