mlsanityData to 14 Oct 2026

Causal Learning and Explanation of Deep Neural Networks via Autoencoded Activations

arXiv 1802.00541 ↗First posted 2 Feb 2018cs.AI cs.LG stat.MLcited by 23

Not on the map. Why?

Abstract (arXiv)

Deep neural networks are complex and opaque. As they enter application in a variety of important and safety critical domains, users seek methods to explain their output predictions. We develop an approach to explaining deep neural networks by constructing causal models on salient concepts contained in a CNN. We develop methods to extract salient concepts throughout a target network by using autoencoders trained to extract human-understandable representations of network activations. We then build a bayesian causal model using these extracted concepts as variables in order to explain image classification. Finally, we use this causal model to identify and visualize features with significant causal influence on final classification.

Key facts. Cited by 23 papers in the map (data through 14 Oct 2026). First posted on arXiv 2 Feb 2018.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2021: 9; 2022: 6; 2023: 4; 2024: 2; 2025: 2.

23 papers.

Sort:

By area:

  1. 2023Causal Abstraction: A Theoretical Foundation for Mechanistic Interpretabilitycited by 205
  2. 2023Recent Advancements in End-to-End Autonomous Driving using Deep Learning: A Surveycited by 103
  3. 2021Explainability of deep vision-based autonomous driving systems: Review and challengescited by 80
  4. 2022Causality-based Neural Network Repaircited by 47
  5. 2021Physics-Guided Deep Learning for Dynamical Systems: A Surveycited by 44
  6. 2021STEEX: Steering Counterfactual Explanations with Semanticscited by 40
  7. 2021Defining and Quantifying the Emergence of Sparse Concepts in DNNscited by 36
  8. 2021Inspect, Understand, Overcome: A Survey of Practical Methods for AI Safetycited by 35
  9. 2022OCTET: Object-aware Counterfactual Explanationscited by 28
  10. 2022Causal Reasoning Meets Visual Representation Learning: A Prospective Studycited by 24
  11. 2022Evaluation Methods and Measures for Causal Learning Algorithmscited by 23
  12. 2024Towards Compositional Interpretability for XAIcited by 12
11 more
  1. 2021Can We Faithfully Represent Masked States to Compute Shapley Values on a DNN?cited by 10
  2. 2021Detecting Multi-Sensor Fusion Errors in Advanced Driver-Assistance Systemscited by 8
  3. 2021Responsible and Regulatory Conform Machine Learning for Medicine: A Survey of Challenges and Solutionscited by 7
  4. 2023Explaining Language Models' Predictions with High-Impact Conceptscited by 7
  5. 2023SUNY: A Visual Interpretation Framework for Convolutional Neural Networks from a Necessary and Sufficient Perspectivecited by 5
  6. 2024GIFT: A Framework Towards Global Interpretable Faithful Textual Explanations of Vision Classifierscited by 2
  7. 2021Explainability-aided Domain Generalization for Image Classificationcited by 1
  8. 2022Explanatory causal effects for model agnostic explanationscited by 1
  9. 2022Neural Bayesian Network Understudycited by 0
  10. 2025Causally-informed Deep Learning towards Explainable and Generalizable Outcomes Prediction in Critical Carecited by 0
  11. 2025A Model of Causal Explanation on Neural Networks for Tabular Datacited by 0