Research map / Bandits and online learning / Contextual bandits and Thompson sampling
PopArt: Efficient Sparse Regression and Experimental Design for Optimal Sparse Linear Bandits
Kyoungseok Jang, Chicheng Zhang, Kwang-Sung Jun
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 PopArt with the H_*^2-optimal sampling design instead of C_min-Lasso with its design | Sparse linear regression · Sparse linear regression, both reported cases | ℓ1 estimation error | Improved |
| Use the PopArt-optimal sampling design with Lasso instead of the C_min-optimal design | Lasso · Sparse linear regression, hard instance where H_*^2 ≪ C_min^{-1} | ℓ1 estimation error | Improved |
| Use PopArt rather than Lasso on the same data points and design | Sparse linear regression · Sparse linear regression, both reported cases | ℓ1 estimation error | Improved |
| Use the PopArt-based bandit algorithm with wPopArt instead of ESTC | Sparse linear bandits · Hard instance and general unit vectors | cumulative regret | Improved |
| Use the H_*^2 experimental-design objective rather than the C_min design objective | PopArt-based sparse linear bandits · Sparse linear bandits, data-poor regime | regret upper bound: H_*^2 ≤ C_min^{-1} ≤ d H_*^2 | Improved |
| Warm up with a coarse estimate and use it as PopArt's pilot estimator | wPopArt · Sparse linear estimation | ℓ1 estimation error bound: 8 s σ √(H^2(Q(μ)) ln(2d/δ)/n_0) | Improved |
Closest papers by meaning
- Sparsity-Agnostic Linear Bandits with Adaptive Adversaries (NeurIPS 2024 · similarity 0.91)
- Causal Bandits: The Pareto Optimal Frontier of Adaptivity, a Reduction to Linear Bandits, and Limitations around Unknown Marginals (ICML 2024 · similarity 0.90)
- Efficient Low-Rank Matrix Estimation, Experimental Design, and Arm-Set-Dependent Low-Rank Bandits (ICML 2024 · similarity 0.90)
- Lasso Bandit with Compatibility Condition on Optimal Arm (ICLR 2025 · similarity 0.90)
- Constrained Linear Thompson Sampling (NeurIPS 2025 · similarity 0.90)
- Improved Regret Bounds for Linear Bandits with Heavy-Tailed Rewards (NeurIPS 2025 · similarity 0.90)
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.
