Research map / Probabilistic and scientific modeling / Treatment effects
A Class of Algorithms for General Instrumental Variable Models
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 the proposed IV bounding method with flexible response functions | General IV model · Synthetic linear Gaussian, weak instrument | causal-effect bound coverage | Helped |
| Use the proposed IV bounding method with flexible response functions | General IV model · Synthetic non-additive, non-linear, strong confounding | causal-effect bound coverage | Helped |
| Use linear response functions instead of a response family capable of representing the true effect (ablation) | Proposed IV bounding method · Synthetic non-additive, non-linear, strong confounding | causal-effect bound coverage | Hurt |
| Increase response-function flexibility from linear toward quadratic and MLP (ablation) | Proposed IV bounding method · Synthetic linear Gaussian | bound tightness | Hurt |
| Use more flexible response functions (ablation) | Proposed IV bounding method · Family Expenditure Survey | bound tightness | Hurt |
| Use an adequate response family with a strong instrument and weak confounding | Proposed IV bounding method · Synthetic non-additive, non-linear, strong instrument | bound tightness | Helped |
| Use linear response functions when the data require a nonlinear response (ablation) | Proposed IV bounding method · Synthetic non-additive, non-linear, strong instrument | optimization feasibility | Hurt |
| Use 2SLS under its additive-model assumptions | 2SLS · Synthetic linear Gaussian | causal-effect accuracy | Helped |
| Apply 2SLS when the outcome is non-additive and nonlinear | 2SLS · Synthetic non-additive, non-linear | causal-effect accuracy | Hurt |
| Apply KIV with a weak instrument | KIV · Synthetic linear Gaussian, weak instrument | causal-effect accuracy | Hurt |
| Apply KIV in the non-additive setting with strong confounding | KIV · Synthetic non-additive, non-linear, strong confounding | causal-effect accuracy | Hurt |
| Apply KIV where the instrument is strong and confounding is weak | KIV · Synthetic non-additive, non-linear, strong instrument | causal-effect accuracy | Helped |
Closest papers by meaning
- Learning Representations of Instruments for Partial Identification of Treatment Effects (ICML 2025 · similarity 0.90)
- Conditional Common Entropy for Instrumental Variable Testing and Partial Identification (ICML 2024 · similarity 0.89)
- Meta-Learners for Partially-Identified Treatment Effects Across Multiple Environments (ICML 2024 · similarity 0.88)
- Towards Estimating Bounds on the Effect of Policies under Unobserved Confounding (NeurIPS 2024 · similarity 0.88)
- Causal Effect Identification in lvLiNGAM from Higher-Order Cumulants (ICML 2025 · similarity 0.87)
- Data Fusion for Partial Identification of Causal Effects (NeurIPS 2025 · similarity 0.87)
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.