Research map / Probabilistic and scientific modeling / Causal discovery
Discovering Causal Signals in Images
Check 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 NCC instead of the previous state-of-the-art causal direction method | Cause-effect direction classification · Tübingen cause-effect datasets v1.0 | accuracy: 75% -> 79% | Helped |
| Select top 1% anticausal features rather than top 1% causal features | NCC feature selection on MSCOCO images · Pascal VOC 2012 object categories in MSCOCO | object-feature ratio | Helped |
| Select top 20% causal and anticausal features instead of top 1% | NCC feature selection on MSCOCO images · Pascal VOC 2012 object categories in MSCOCO | classes with higher object-feature ratio for anticausal features: 16 out of 20 classes | Mixed |
| Replace NCC scores with the correlation coefficient or its absolute value | Feature selection on MSCOCO images · Pascal VOC 2012 object categories in MSCOCO | relationship between anticausal features and object features | Hurt |
| Add randomly permuted, uncorrelated examples labeled 1/2 to NCC training | NCC for image feature analysis · MSCOCO image feature and category scores | causal feature classification | No clear effect |
Closest papers by meaning
- A Sparsity Principle for Partially Observable Causal Representation Learning (ICML 2024 · similarity 0.87)
- Learning Linear Causal Representations from General Environments: Identifiability and Intrinsic Ambiguity (NeurIPS 2024 · similarity 0.86)
- Bayesian Active Learning for Bivariate Causal Discovery (ICML 2025 · similarity 0.86)
- Multi-View Causal Representation Learning with Partial Observability (ICLR 2024 · similarity 0.86)
- Causal Representation Learning Made Identifiable by Grouping of Observational Variables (ICML 2024 · similarity 0.86)
- Object centric architectures enable efficient causal representation learning (ICLR 2024 · similarity 0.86)
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.