Research map / Probabilistic and scientific modeling / Treatment effects
Quantifying Ignorance in Individual-Level Causal-Effect Estimates under Hidden Confounding
Andrew Jesson, Sören Mindermann, Yarin Gal, Uri Shalit
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 the proposed parametric interval estimator instead of the baseline at matched Γ | Policy estimation on simulated data · Simulated data, log Γ*=0.5 | policy risk error: 0.10±0.04 -> 0.07±0.03 | Improved |
| Use the proposed parametric interval estimator instead of the baseline at matched Γ | Policy estimation on simulated data · Simulated data, log Γ*=1.0 | policy risk error: 0.25±0.11 -> 0.10±0.04 | Improved |
| Use the proposed parametric interval estimator instead of the baseline at matched Γ | Policy estimation on simulated data · Simulated data, log Γ*=1.5 | policy risk error: 0.52±0.40 -> 0.13±0.04 | Improved |
| Match the assumed Γ to the true hidden-confounding level Γ* | Proposed method · Simulated data | policy risk error: 0.38±0.04 -> 0.07±0.03; 0.31±0.03 -> 0.10±0.04; 0.75±0.18 -> 0.13±0.04 | Improved |
| Use the proposed interval estimator on high-dimensional MNIST data | Policy estimation · HC-MNIST | policy risk error | Improved |
| Account jointly for epistemic uncertainty and hidden-confounding sensitivity rather than either alone | IHDP deferral policy · IHDP Hidden Confounding | recommendation error rate | Improved |
| Add epistemic uncertainty to sensitivity-based intervals (ablation) | IHDP deferral policy · IHDP Hidden Confounding | recommendation error rate | Improved |
| Add hidden-confounding sensitivity to epistemic-uncertainty-based intervals (ablation) | IHDP deferral policy · IHDP Hidden Confounding | recommendation error rate | Improved |
Closest papers by meaning
- A Non-parametric Direct Learning Approach to Heterogeneous Treatment Effect Estimation under Unmeasured Confounding (NeurIPS 2024 · similarity 0.91)
- Impossibility of Distribution-Free Predictive Inference for Individual Treatment Effects (NeurIPS 2026 · similarity 0.89)
- Learning Representations of Instruments for Partial Identification of Treatment Effects (ICML 2025 · similarity 0.89)
- Do Contemporary Causal Inference Models Capture Real-World Heterogeneity? Findings from a Large-Scale Benchmark (ICLR 2025 · similarity 0.89)
- Quantifying Aleatoric Uncertainty of the Treatment Effect: A Novel Orthogonal Learner (NeurIPS 2024 · similarity 0.89)
- Bounds on Representation-Induced Confounding Bias for Treatment Effect Estimation (ICLR 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.
