mlsanityData to 14 Oct 2026

Explaining Classifiers with Causal Concept Effect (CaCE)

arXiv 1907.07165 ↗First posted 16 Jul 2019cs.LG cs.CV stat.MLcited by 135

Not on the map. Why?

Abstract (arXiv)

How can we understand classification decisions made by deep neural networks? Many existing explainability methods rely solely on correlations and fail to account for confounding, which may result in potentially misleading explanations. To overcome this problem, we define the Causal Concept Effect (CaCE) as the causal effect of (the presence or absence of) a human-interpretable concept on a deep neural net's predictions. We show that the CaCE measure can avoid errors stemming from confounding. Estimating CaCE is difficult in situations where we cannot easily simulate the do-operator. To mitigate this problem, we use a generative model, specifically a Variational AutoEncoder (VAE), to measure VAE-CaCE. In an extensive experimental analysis, we show that the VAE-CaCE is able to estimate the true concept causal effect, compared to baselines for a number of datasets including high dimensional images.

Key facts. Cited by 135 papers in the map (data through 14 Oct 2026). First posted on arXiv 16 Jul 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 27; 2022: 30; 2023: 19; 2024: 27; 2025: 19; 2026: 13.

135 papers.

Sort:

By area:

  1. 2022Post-hoc Concept Bottleneck Modelscited by 293
  2. 2021Post-hoc Interpretability for Neural NLP: A Surveycited by 156
  3. 2021Benchmarking and Survey of Explanation Methods for Black Box Modelscited by 136
  4. 2021Acquisition of Chess Knowledge in AlphaZerocited by 123
  5. 2021Explaining in Style: Training a GAN to explain a classifier in StyleSpacecited by 122
  6. 2021Counterfactual Generative Networkscited by 111
  7. 2023Concept-based Explainable Artificial Intelligence: A Surveycited by 98
  8. 2021CARLA: A Python Library to Benchmark Algorithmic Recourse and Counterfactual Explanation Algorithmscited by 86
  9. 2021Leveraging Sparse Linear Layers for Debuggable Deep Networkscited by 82
  10. 2022Concept Bottleneck Model with Additional Unsupervised Conceptscited by 79
  11. 2021Editing a classifier by rewriting its prediction rulescited by 65
  12. 2022CEBaB: Estimating the Causal Effects of Real-World Concepts on NLP Model Behaviorcited by 55
38 more
  1. 2021Entropy-based Logic Explanations of Neural Networkscited by 51
  2. 2022A Causal Lens for Controllable Text Generationcited by 45
  3. 2023Faithful Explanations of Black-box NLP Models Using LLM-generated Counterfactualscited by 44
  4. 2021ELLA: Exploration through Learned Language Abstractioncited by 42
  5. 2023Bridging the Human-AI Knowledge Gap: Concept Discovery and Transfer in AlphaZerocited by 42
  6. 2023Learning to Receive Help: Intervention-Aware Concept Embedding Modelscited by 41
  7. 2021What if This Modified That? Syntactic Interventions via Counterfactual Embeddingscited by 39
  8. 2021DISSECT: Disentangled Simultaneous Explanations via Concept Traversalscited by 37
  9. 2023Explain Any Concept: Segment Anything Meets Concept-Based Explanationcited by 37
  10. 2022Cycle-Consistent Counterfactuals by Latent Transformationscited by 33
  11. 2023Interpreting Pretrained Language Models via Concept Bottleneckscited by 33
  12. 2022Causal Proxy Models for Concept-Based Model Explanationscited by 29
  13. 2024Causal Concept Graph Models: Beyond Causal Opacity in Deep Learningcited by 29
  14. 2022Explainable AI for clinical and remote health applications: a survey on tabular and time series datacited by 24
  15. 2023A Glitch in the Matrix? Locating and Detecting Language Model Grounding with Fakepediacited by 24
  16. 2024The Quest for the Right Mediator: Surveying Mechanistic Interpretability Through the Lens of Causal Mediation Analysiscited by 24
  17. 2022Evaluation Methods and Measures for Causal Learning Algorithmscited by 23
  18. 2022Human-Centered Concept Explanations for Neural Networkscited by 23
  19. 2022Unit Testing for Concepts in Neural Networkscited by 23
  20. 2021Best of both worlds: local and global explanations with human-understandable conceptscited by 22
  21. 2021On Causally Disentangled Representationscited by 22
  22. 2022Understanding and Enhancing Robustness of Concept-based Modelscited by 22
  23. 2022Concept Embedding Analysis: A Reviewcited by 20
  24. 2022Concept Gradient: Concept-based Interpretation Without Linear Assumptioncited by 20
  25. 2024Counterfactual Image Editingcited by 20
  26. 2021Rationalization through Conceptscited by 19
  27. 2022Unpacking Large Language Models with Conceptual Consistencycited by 18
  28. 2024Concept-Based Explanations in Computer Vision: Where Are We and Where Could We Go?cited by 18
  29. 2025Causally Reliable Concept Bottleneck Modelscited by 17
  30. 2022ECLAD: Extracting Concepts with Local Aggregated Descriptorscited by 15
  31. 2023The Co-12 Recipe for Evaluating Interpretable Part-Prototype Image Classifierscited by 15
  32. 2021Model-Based Counterfactual Synthesizer for Interpretationcited by 14
  33. 2021Learn-Explain-Reinforce: Counterfactual Reasoning and Its Guidance to Reinforce an Alzheimer's Disease Diagnosis Modelcited by 14
  34. 2021Latent Space Explanation by Interventioncited by 13
  35. 2022Do Users Benefit From Interpretable Vision? A User Study, Baseline, And Datasetcited by 13
  36. 2022Spatial-temporal Concept based Explanation of 3D ConvNetscited by 12
  37. 2022Adapting to Latent Subgroup Shifts via Concepts and Proxiescited by 12
  38. 2023Causal reasoning in typical computer vision taskscited by 12

50 most cited of 135 All 135 (JSON)