Challenges in Variable Importance Ranking Under Correlation

• URL: http://arxiv.org/abs/2402.03447v1
• Date: Mon, 5 Feb 2024 19:02:13 GMT
• Title: Challenges in Variable Importance Ranking Under Correlation
• Authors: Annie Liang and Thomas Jemielita and Andy Liaw and Vladimir Svetnik and Lingkang Huang and Richard Baumgartner and Jason M. Klusowski
• Abstract summary: We present a comprehensive simulation study investigating the impact of feature correlation on the assessment of variable importance. While there is always no correlation between knockoff variables and its corresponding predictor variables, we prove that the correlation increases linearly beyond a certain correlation threshold between the predictor variables.
• Score: 6.718144470265263
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Variable importance plays a pivotal role in interpretable machine learning as it helps measure the impact of factors on the output of the prediction model. Model agnostic methods based on the generation of "null" features via permutation (or related approaches) can be applied. Such analysis is often utilized in pharmaceutical applications due to its ability to interpret black-box models, including tree-based ensembles. A major challenge and significant confounder in variable importance estimation however is the presence of between-feature correlation. Recently, several adjustments to marginal permutation utilizing feature knockoffs were proposed to address this issue, such as the variable importance measure known as conditional predictive impact (CPI). Assessment and evaluation of such approaches is the focus of our work. We first present a comprehensive simulation study investigating the impact of feature correlation on the assessment of variable importance. We then theoretically prove the limitation that highly correlated features pose for the CPI through the knockoff construction. While we expect that there is always no correlation between knockoff variables and its corresponding predictor variables, we prove that the correlation increases linearly beyond a certain correlation threshold between the predictor variables. Our findings emphasize the absence of free lunch when dealing with high feature correlation, as well as the necessity of understanding the utility and limitations behind methods in variable importance estimation.

Related papers

• Nonparametric Partial Disentanglement via Mechanism Sparsity: Sparse Actions, Interventions and Sparse Temporal Dependencies [58.179981892921056]
  This work introduces a novel principle for disentanglement we call mechanism sparsity regularization. We propose a representation learning method that induces disentanglement by simultaneously learning the latent factors. We show that the latent factors can be recovered by regularizing the learned causal graph to be sparse.
  arXiv  Detail & Related papers  (2024-01-10T02:38:21Z)
• Identifiable Latent Polynomial Causal Models Through the Lens of Change [82.14087963690561]
  Causal representation learning aims to unveil latent high-level causal representations from observed low-level data. One of its primary tasks is to provide reliable assurance of identifying these latent causal models, known as identifiability.
  arXiv  Detail & Related papers  (2023-10-24T07:46:10Z)
• A Notion of Feature Importance by Decorrelation and Detection of Trends by Random Forest Regression [1.675857332621569]
  We introduce a novel notion of feature importance based on the well-studied Gram-Schmidt decorrelation method. We propose two estimators for identifying trends in the data using random forest regression.
  arXiv  Detail & Related papers  (2023-03-02T11:01:49Z)
• Data-Driven Influence Functions for Optimization-Based Causal Inference [105.5385525290466]
  We study a constructive algorithm that approximates Gateaux derivatives for statistical functionals by finite differencing. We study the case where probability distributions are not known a priori but need to be estimated from data.
  arXiv  Detail & Related papers  (2022-08-29T16:16:22Z)
• Decorrelated Variable Importance [0.0]
  We propose a method for mitigating the effect of correlation by defining a modified version of LOCO. This new parameter is difficult to estimate nonparametrically, but we show how to estimate it using semiparametric models.
  arXiv  Detail & Related papers  (2021-11-21T16:31:36Z)
• Decoding Causality by Fictitious VAR Modeling [0.0]
  We first set up an equilibrium for the cause-effect relations using a fictitious vector autoregressive model. In the equilibrium, long-run relations are identified from noise, and spurious ones are negligibly close to zero. We also apply the approach to estimating the causal factors' contribution to climate change.
  arXiv  Detail & Related papers  (2021-11-14T22:43:02Z)
• Variational Causal Networks: Approximate Bayesian Inference over Causal Structures [132.74509389517203]
  We introduce a parametric variational family modelled by an autoregressive distribution over the space of discrete DAGs. In experiments, we demonstrate that the proposed variational posterior is able to provide a good approximation of the true posterior.
  arXiv  Detail & Related papers  (2021-06-14T17:52:49Z)
• Latent Causal Invariant Model [128.7508609492542]
  Current supervised learning can learn spurious correlation during the data-fitting process. We propose a Latent Causal Invariance Model (LaCIM) which pursues causal prediction.
  arXiv  Detail & Related papers  (2020-11-04T10:00:27Z)
• Estimating Causal Effects with the Neural Autoregressive Density Estimator [6.59529078336196]
  We use neural autoregressive density estimators to estimate causal effects within the Pearl's do-calculus framework. We show that the approach can retrieve causal effects from non-linear systems without explicitly modeling the interactions between the variables.
  arXiv  Detail & Related papers  (2020-08-17T13:12:38Z)
• On Disentangled Representations Learned From Correlated Data [59.41587388303554]
  We bridge the gap to real-world scenarios by analyzing the behavior of the most prominent disentanglement approaches on correlated data. We show that systematically induced correlations in the dataset are being learned and reflected in the latent representations. We also demonstrate how to resolve these latent correlations, either using weak supervision during training or by post-hoc correcting a pre-trained model with a small number of labels.
  arXiv  Detail & Related papers  (2020-06-14T12:47:34Z)

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.