mlsanityData to 7 Oct 2026

Instance-wise Causal Feature Selection for Model Interpretation

arXiv 2104.12759 ↗First posted 26 Apr 2021cs.LG cs.AIcited by 8

Not on the map. Why?

Abstract (arXiv)

We formulate a causal extension to the recently introduced paradigm of instance-wise feature selection to explain black-box visual classifiers. Our method selects a subset of input features that has the greatest causal effect on the models output. We quantify the causal influence of a subset of features by the Relative Entropy Distance measure. Under certain assumptions this is equivalent to the conditional mutual information between the selected subset and the output variable. The resulting causal selections are sparser and cover salient objects in the scene. We show the efficacy of our approach on multiple vision datasets by measuring the post-hoc accuracy and Average Causal Effect of selected features on the models output.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 18 of its 21 references that are in the map. First posted on arXiv 26 Apr 2021.

Builds on

18 of 21 references.

References · 18

  1. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  2. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  3. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  4. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,708
  5. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
  6. 2018Sanity Checks for Saliency Mapscited by 1,149
  7. 2016The Concrete Distribution: A Continuous Relaxation of Discrete Random Variablescited by 1,135
  8. 2017SmoothGrad: removing noise by adding noisecited by 1,097
  9. 2020Inverting Gradients -- How easy is it to break privacy in federated learning?cited by 707
  10. 2018Learning to Explain: An Information-Theoretic Perspective on Model Interpretationcited by 248
  11. 2019Explainable Machine Learning in Deploymentcited by 210
  12. 2020Problems with Shapley-value-based explanations as feature importance measurescited by 198
6 more
  1. 2020Invariant Rationalizationcited by 131
  2. 2018L-Shapley and C-Shapley: Efficient Model Interpretation for Structured Datacited by 100
  3. 2019CXPlain: Causal Explanations for Model Interpretation under Uncertaintycited by 98
  4. 2019Neural Network Attributions: A Causal Perspectivecited by 77
  5. 2020Generative causal explanations of black-box classifierscited by 41
  6. 2018Granger-causal Attentive Mixtures of Experts: Learning Important Features with Neural Networkscited by 13

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 0; 2024: 3; 2025: 1; 2026: 1.

8 papers.

  1. 2022Deep Causal Learning: Representation, Discovery and Inferencecited by 11
  2. 2022DIWIFT: Discovering Instance-wise Influential Features for Tabular Datacited by 6
  3. 2024Partial Information Decomposition for Data Interpretability and Feature Selectioncited by 3
  4. 2024Causal Feature Selection for Responsible Machine Learningcited by 2
  5. 2024Causal Prototype-inspired Contrast Adaptation for Unsupervised Domain Adaptive Semantic Segmentation of High-resolution Remote Sensing Imagerycited by 1
  6. 2022Causality for Inherently Explainable Transformers: CAT-XPLAINcited by 0
  7. 2025Unsupervised Dynamic Feature Selection for Robust Latent Spaces in Vision Taskscited by 0
  8. 2026Explaining Black-Box Language Models: Learning to Optimize Linguistically-Structured Word Subsetscited by 0