Research map / Probabilistic and scientific modeling / Causal discovery
Neural Network Attributions: A Causal Perspective
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use second-order directional derivatives instead of exact Hessian-based calculation | ACE approximation on MNIST with a 4-hidden-layer DNN · MNIST | computation time per computation: 3.04s -> 0.04s | Helped |
| Impute the first sequence feature x^0 | trained GRU · Simulated sequence classification, 1M test sequences | prediction changes: 1956 | Mixed |
| Impute the second sequence feature x^1 | trained GRU · Simulated sequence classification, 1M test sequences | prediction changes: 9 | Mixed |
| Impute the third sequence feature x^3 | trained GRU · Simulated sequence classification, 1M test sequences | prediction changes | No clear effect |
| Use causal attributions instead of Integrated Gradients to explain the recurrent classifier | trained GRU · Simulated sequence classification | identification of features used by the learned model | Helped |
Closest papers by meaning
- Towards Learning and Explaining Indirect Causal Effects in Neural Networks (AAAI 2024 · similarity 0.90)
- Rethinking Causal Relationships Learning in Graph Neural Networks (AAAI 2024 · similarity 0.86)
- The Non-Linear Representation Dilemma: Is Causal Abstraction Enough for Mechanistic Interpretability? (NeurIPS 2025 · similarity 0.86)
- Since Faithfulness Fails: The Performance Limits of Neural Causal Discovery (ICML 2025 · similarity 0.86)
- Neural Causal Abstractions (AAAI 2024 · similarity 0.85)
- Causal Concept Graph Models: Beyond Causal Opacity in Deep Learning (ICLR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.