Research map / Probabilistic and scientific modeling / Molecules and proteins
Sparse Recovery With Multiple Data Streams: An Adaptive Sequential Testing Approach
Weinan Wang, Bowen Gang, Wenguang Sun
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 |
|---|---|---|---|
| replace simple single thresholding (parallel SPRTs) with compound thresholding (SMART) (ablation) | sequential multiple testing, p=10^5 · Simulated mixture model Settings 1-3 | FPR/MDR control and expected sample size: FPR as low as 0.01 when the nominal FPR level is 0.05 | Improved |
| use estimated parameters instead of oracle parameters in SMART (ablation) | SMART · Sparse and weak signals, pi=0.01 | MDR | Hurt |
| use SMART instead of distilled sensing at matched error rates | distilled sensing (up to 10 stages) · Simulated mixture model, p=10^5 | expected sample size | Improved |
| use SMART instead of the two-stage Hunt procedure | Hunt (2 stages, 2x10^5 samples) · Simulated mixture model, nominal FDR and MDR 0.05 | MDR control and sample size | Improved |
| add correlated errors (Gaussian process, tridiagonal, AR) for SMART (ablation) | DD.SM · Dependent tests, p=10^4 | FPR/MDR | Mixed |
Closest papers by meaning
- A Bandit Approach to Sequential Experimental Design with False Discovery Control (NeurIPS 2018 · similarity 0.88)
- ADDIS: an adaptive discarding algorithm for online FDR control with conservative nulls (NeurIPS 2019 · similarity 0.86)
- Anytime-Valid PAC-Bayes Certificates for Adaptive Test-Time Scaling (NeurIPS 2026 · similarity 0.86)
- Adaptive Variance Inflation in Thompson Sampling: Efficiency, Safety, Robustness, and Beyond (NeurIPS 2025 · similarity 0.86)
- Provable Posterior Sampling with Denoising Oracles via Tilted Transport (NeurIPS 2024 · similarity 0.85)
- Straight-Through Meets Sparse Recovery: the Support Exploration Algorithm (ICML 2024 · similarity 0.84)
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.
