Research map / Learning theory and algorithms / Prediction and decision losses
Treatment Responder Classification with Abstention
Haoxiang Wang, Aoqi Zuo, Ziyan Wang, Zhiheng Zhang, Erdun Gao, Kun Zhang, Haoxuan Li, Mingming Gong
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 |
|---|---|---|---|
| relax unconfoundedness (w/o uncf.) instead of full TRECA (ablation) | TRECA · Twins mono | f-regret-r: 0.001 -> 0.004 | Hurt |
| add monotonicity assumption (ablation) | TRECA · Twins mono | f-regret-r: 0.002 -> 0.001 | Improved |
| add monotonicity assumption (ablation) | TRECA · Jobs mono | f-regret-r: 0.046 -> 0.017 | Improved |
| add unconfoundedness-based estimation (vs w/o uncf.) (ablation) | TRECA · Jobs mono | f-regret-r: 0.039 -> 0.017 | Improved |
| use monotonicity-free variant when monotonicity is violated (ablation) | TRECA · Twins | f-regret-r: 0.014 -> 0.006 | Hurt |
| use monotonicity-free variant when monotonicity is violated (ablation) | TRECA · Jobs | f-regret-r: 0.069 -> 0.036 | Hurt |
| misspecify propensity (fix to 0.25) (ablation) | TRECA · Twins mono | f-regret-r: 0.001 -> 0.021 | Hurt |
| misspecify propensity (fix to 0.25) (ablation) | TRECA · Jobs mono | f-regret-r: 0.017 -> 0.035 | Hurt |
| replace CFRNet with TRECA | TRECA · Twins mono | f-regret-r: 0.018 -> 0.001 | Improved |
| replace CFRNet with TRECA | TRECA · Jobs mono | f-regret-r: 0.039 -> 0.017 | Improved |
| replace CFRNet with TRECA | TRECA · Twins mono | f-regret-a: 0.063 -> 0.352 | Hurt |
| replace CFRNet with TRECA | TRECA · Jobs mono | f-regret-a: 0.082 -> 0.167 | Hurt |
Closest papers by meaning
- Quantifying Aleatoric Uncertainty of the Treatment Effect: A Novel Orthogonal Learner (NeurIPS 2024 · similarity 0.86)
- Assessing the robustness of heterogeneous treatment effects in survival analysis under informative censoring (NeurIPS 2026 · similarity 0.84)
- Towards Safe Policy Learning under Partial Identifiability: A Causal Approach (AAAI 2024 · similarity 0.84)
- Doubly robust identification of treatment effects from multiple environments (ICLR 2025 · similarity 0.84)
- Meta-Learners for Partially-Identified Treatment Effects Across Multiple Environments (ICML 2024 · similarity 0.83)
- An effective framework for estimating individualized treatment rules (NeurIPS 2024 · similarity 0.83)
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.
