Research map / Probabilistic and scientific modeling / Treatment effects
Treatment Effect Estimation with Disentangled Latent Factors
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 TEDVAE's disentangled latent factors instead of X-RF | Treatment effect estimation · ACIC2016 | test PEHE: 1.89±0.16 -> 1.77±0.17 | Helped |
| Use TEDVAE's disentangled latent factors instead of DR-CFR | Treatment effect estimation · IHDP Setting A | test PEHE: 0.65±0.18 -> 0.60±0.14 | Helped |
| Use TEDVAE's disentangled latent factors instead of CFR | Treatment effect estimation · IHDP Setting B | test PEHE: 2.76±0.04 -> 2.22±0.08 | Helped |
| Use TEDVAE's disentangled latent factors instead of GANITE | Treatment effect estimation · Twins | test ATE error: 0.018±0.005 -> 0.006±0.002 | Helped |
| Add instrumental factors rather than force their latent dimension to zero (ablation) | TEDVAE · Synthetic datasets | PEHE | Helped |
| Add confounding factors rather than force their latent dimension to zero (ablation) | TEDVAE · Synthetic datasets | PEHE | Helped |
| Add risk factors rather than force their latent dimension to zero (ablation) | TEDVAE · Synthetic datasets | PEHE | Helped |
| Disentangle instrumental, confounding, and risk factors rather than treat all observed variables as confounders | TEDVAE versus CEVAE · Synthetic data with instrumental and risk factors | PEHE | Helped |
| Use TEDVAE's inference procedure when instrumental and risk factors are absent | TEDVAE versus CEVAE · Synthetic data without instrumental and risk factors | PEHE | Mixed |
Closest papers by meaning
- A Non-parametric Direct Learning Approach to Heterogeneous Treatment Effect Estimation under Unmeasured Confounding (NeurIPS 2024 · similarity 0.90)
- Estimating Heterogeneous Treatment Effects by Combining Weak Instruments and Observational Data (NeurIPS 2024 · similarity 0.89)
- An Analysis of Causal Effect Estimation using Outcome Invariant Data Augmentation (NeurIPS 2025 · similarity 0.88)
- Meta-Learners for Partially-Identified Treatment Effects Across Multiple Environments (ICML 2024 · similarity 0.87)
- Automating the Selection of Proxy Variables of Unmeasured Confounders (ICML 2024 · similarity 0.87)
- Learning Representations of Instruments for Partial Identification of Treatment Effects (ICML 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.