Research map / Bandits and online learning / Online convex optimization and games
A Bandit Framework for Strategic Regression
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 SR-UCB (UCB-style index with Brier scoring rule, per-round base payment) to select workers instead of a static payment rule | dynamic data acquisition for linear regression · theoretical analysis (linear regression) | approximate BNE gap | Helped |
| Add randomized task assignment (probability p_s when multiple workers selected) to remove the all-shirk equilibrium | SR-UCB for linear regression · theoretical analysis (linear regression) | equilibrium existence of all-shirking | Helped |
| Use online updates for the regression estimator and scores (OSR1/OSR2-UCB) instead of recomputing estimators each round | SR-UCB · theoretical analysis (ridge / online updates) | BNE gap: O(sqrt(logT/T)) to O(logT/sqrt(T)) for OSR2-UCB | Hurt |
| Add differential privacy noise (PSR-UCB) to model and indexes | SR-UCB · theoretical analysis (linear regression) | BNE gap and privacy level: BNE gap O(sqrt(logT/T)) to O(log^6 T/sqrt(T)); achieves (O(log^-1 T), O(log^-1 T))-DP | Hurt |
Closest papers by meaning
- Robustly Improving Bandit Algorithms with Confounded and Selection Biased Offline Data: A Causal Approach (AAAI 2024 · similarity 0.86)
- Leveraging (Biased) Information: Multi-armed Bandits with Offline Data (ICML 2024 · similarity 0.85)
- Collaborative Mean Estimation Among Heterogeneous Strategic Agents: Individual Rationality, Fairness, and Truthful Contribution (ICML 2025 · similarity 0.85)
- Locally Private and Robust Multi-Armed Bandits (NeurIPS 2024 · similarity 0.85)
- Truthful High Dimensional Sparse Linear Regression (NeurIPS 2024 · similarity 0.85)
- A Cramér–von Mises Approach to Incentivizing Truthful Data Sharing (NeurIPS 2025 · 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.