Research map / Reinforcement learning / Policy gradient and actor-critic
Risk-Averse Active Sensing for Timely Outcome Prediction under Cost Pressure
Yuchao Qin, Mihaela van der Schaar, Changhee Lee
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 risk-averse tail objective with adaptive intervals instead of constant-interval AS baseline | RAS · synthetic DS | timeliness d(delta=0.5): 0.522±0.012 -> 0.264±0.086 | Improved |
| use RAS instead of NLL adaptive sensing | RAS · synthetic DS | timeliness d(delta=0.5): 0.974±0.141 -> 0.264±0.086 | Improved |
| use RAS instead of ASAC | RAS · synthetic DS | ROC AUC: 0.605±0.096 -> 0.680±0.003 | Improved |
| use RAS instead of dense-sensing predictor FO | RAS · synthetic DS | timeliness d(delta=0.5): 0.349±0.000 -> 0.264±0.086 | Improved |
| use RAS instead of dense sensing FO | RAS · synthetic DS | acquisition cost: 31.000±0.000 -> 6.077±0.953 | Improved |
| focus on tail set (risk-averse) instead of risk-neutral AS | RAS · synthetic DS | acquisition cost: 4.501±0.497 -> 6.077±0.953 | Hurt |
| use RAS instead of AS baseline | RAS · ADNI | ROC AUC: 0.704±0.023 -> 0.730±0.007 | Improved |
| use RAS instead of AS baseline | RAS · ADNI | timeliness d(delta=0.5): 2.357±0.375 -> 1.192±0.176 | Improved |
| use RAS instead of NLL | RAS · ADNI | timeliness d(delta=0.1): 1.040±0.149 -> 0.820±0.096 | Improved |
| use RAS instead of dense sensing FO | RAS · ADNI | timeliness d(delta=0.1): 0.141±0.000 -> 0.820±0.096 | Hurt |
| use RAS instead of dense sensing FO | RAS · ADNI | ROC AUC: 0.747±0.000 -> 0.730±0.007 | Hurt |
| use RAS instead of dense sensing FO | RAS · ADNI | acquisition cost: 26.865±0.000 -> 8.614±1.157 | Improved |
Closest papers by meaning
- Decision Theoretic Foundations for Conformal Prediction: Optimal Uncertainty Quantification for Risk-Averse Agents (ICML 2025 · similarity 0.86)
- Enhancing Treatment Effect Estimation via Active Learning: A Counterfactual Covering Perspective (ICML 2025 · similarity 0.85)
- Active, anytime-valid risk controlling prediction sets (NeurIPS 2024 · similarity 0.84)
- Orthogonal Survival Learners for Estimating Heterogeneous Treatment Effects from Time-to-Event Data (NeurIPS 2025 · similarity 0.83)
- DARTS: Targeting Prognostic Covariates in Budget-Constrained Sequential Experiments (NeurIPS 2026 · similarity 0.83)
- Risk Horizons: Structured Hypothesis Spaces for Longitudinal Clinical Prediction (NeurIPS 2026 · 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.
