Research map / Probabilistic and scientific modeling / Treatment effects
Learning Instrumental Variable from Data Fusion for Treatment Effect Estimation
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 |
|---|---|---|---|
| Reconstruct the latent group IV with Meta-EM instead of using no IV | AGMM · Linear-3-3 | MSE: 0.214(0.046) -> 0.064(0.004) | Helped |
| Replace K-means group reconstruction with EM group reconstruction (ablation) | AGMM with group IV · Linear-3-3 | MSE: 0.123(0.025) -> 0.064(0.004) | Helped |
| Reconstruct the latent group IV with Meta-EM instead of using no IV | AGMM · IHDP | MSE: 0.140(0.063) -> 0.095(0.035) | Helped |
| Replace K-means group reconstruction with EM group reconstruction (ablation) | AGMM with group IV · IHDP | MSE: 0.112(0.050) -> 0.095(0.035) | Helped |
| Reconstruct the latent group IV with Meta-EM instead of using no IV | AGMM · PM-CMR | MSE: 0.130(0.064) -> 0.085(0.045) | Helped |
| Replace K-means group reconstruction with EM group reconstruction (ablation) | AGMM with group IV · PM-CMR | MSE: 0.117(0.053) -> 0.085(0.045) | Helped |
| Add a shared learned representation and alternate representation learning with EM (ablation) | EM group reconstruction · Data-2-5, linear | Reconstruction accuracy: 86.1% -> 86.1% | No clear effect |
| Add a shared learned representation and alternate representation learning with EM (ablation) | EM group reconstruction · Data-2-5, poly | Reconstruction accuracy: 82.9% -> 92.7% | Helped |
| Add a shared learned representation and alternate representation learning with EM (ablation) | EM group reconstruction · Data-2-5, sin | Reconstruction accuracy: 87.2% -> 90.3% | Helped |
| Add a shared learned representation and alternate representation learning with EM (ablation) | EM group reconstruction · Data-2-5, sigmoid | Reconstruction accuracy: 89.4% -> 94.7% | Helped |
| Add a shared learned representation and alternate representation learning with EM (ablation) | EM group reconstruction · Data-2-5, abs | Reconstruction accuracy: 85.2% -> 92.3% | Helped |
| Use learned nonlinear representations rather than known nonlinear treatment functions (ablation) | Group IV reconstruction · Data-2-5, poly | Reconstruction accuracy: 98.1% -> 92.7% | Hurt |
Closest papers by meaning
- Meta-Learners for Partially-Identified Treatment Effects Across Multiple Environments (ICML 2024 · similarity 0.89)
- Estimating Heterogeneous Treatment Effects by Combining Weak Instruments and Observational Data (NeurIPS 2024 · similarity 0.88)
- An Analysis of Causal Effect Estimation using Outcome Invariant Data Augmentation (NeurIPS 2025 · similarity 0.88)
- A Non-parametric Direct Learning Approach to Heterogeneous Treatment Effect Estimation under Unmeasured Confounding (NeurIPS 2024 · similarity 0.87)
- Hierarchical and Density-based Causal Clustering (NeurIPS 2024 · similarity 0.86)
- Learning Representations of Instruments for Partial Identification of Treatment Effects (ICML 2025 · 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.