Research map / Probabilistic and scientific modeling / Causal discovery
Causal Discovery with Latent Confounders Based on Higher-Order Cumulants
Ruichu Cai, Zhiyi Huang, Wei Chen, Zhifeng Hao, Kun Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 higher-order cumulants and the One-Latent-Component condition for causal discovery | Canonical lvLiNGAM · Synthetic Case 2, N=1000 | directed-edge F1: 0.50 (0.00) -> 0.70 (0.21) | Improved |
| Use higher-order cumulants and the One-Latent-Component condition for causal discovery | Canonical lvLiNGAM · Synthetic Case 3, N=1000 | directed-edge F1: 0.45 (0.01) -> 0.40 (0.11) | Hurt |
| Use higher-order cumulants and the One-Latent-Component condition for causal discovery | Canonical lvLiNGAM · Synthetic Case 4, N=1000 | directed-edge F1: 0.48 (0.01) -> 0.49 (0.12) | Improved |
| Use higher-order cumulants and the One-Latent-Component condition for causal discovery | Canonical lvLiNGAM · Synthetic Case 6, N=1000 | directed-edge F1: 0.17 (0.02) -> 0.40 (0.24) | Improved |
| Use iterative identification and surrogate regression to discover non-adjacent relations | Canonical lvLiNGAM · Synthetic Case 1, N=1000 | non-adjacent-relationship F1: 0.35 (0.05) -> 0.94 (0.01) | Improved |
| Use iterative identification and surrogate regression to discover non-adjacent relations | Canonical lvLiNGAM · Synthetic Case 2, N=1000 | non-adjacent-relationship F1: 0.00 (0.00) -> 1.00 (0.00) | Improved |
| Use iterative identification and surrogate regression to discover non-adjacent relations | Canonical lvLiNGAM · Synthetic Case 3, N=1000 | non-adjacent-relationship F1: 0.69 (0.00) -> 0.83 (0.03) | Improved |
| Use iterative identification and surrogate regression to discover non-adjacent relations | Canonical lvLiNGAM · Synthetic Case 4, N=1000 | non-adjacent-relationship F1: 0.48 (0.01) -> 0.86 (0.00) | Improved |
| Use iterative identification and surrogate regression to discover non-adjacent relations | Canonical lvLiNGAM · Synthetic Case 5, N=1000 | non-adjacent-relationship F1: 0.31 (0.00) -> 0.80 (0.02) | Improved |
| Use iterative identification and surrogate regression to discover non-adjacent relations | Canonical lvLiNGAM · Synthetic Case 6, N=1000 | non-adjacent-relationship F1: 0.22 (0.04) -> 0.98 (0.01) | Improved |
Closest papers by meaning
- Identification of Causal Structure with Latent Variables Based on Higher Order Cumulants (AAAI 2024 · similarity 0.94)
- Differentiable Causal Discovery for Latent Hierarchical Causal Models (ICLR 2025 · similarity 0.90)
- Identification of Latent Confounders via Investigating the Tensor Ranks of the Nonlinear Observations (ICML 2025 · similarity 0.89)
- A Versatile Causal Discovery Framework to Allow Causally-Related Hidden Variables (ICLR 2024 · similarity 0.89)
- Latent Variable Causal Discovery under Selection Bias (ICML 2025 · similarity 0.88)
- Learning Discrete Latent Variable Structures with Tensor Rank Conditions (NeurIPS 2024 · similarity 0.88)
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.
