Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Scalable Generalized Linear Bandits: Online Computation and Hashing
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 GLOC instead of UCB-GLM or OL2M | GLB algorithms · Logit model | cumulative regret | No clear effect |
| Use OL2M, which is designed for the logit model, on probit rewards | GLB algorithms · Probit model | cumulative regret | Hurt |
| Use hashing to evaluate the arm-selection objective on 1% of the arms | QGLOC · Logit model, N=100,000, T=5000 | cumulative regret | Hurt |
| Use hashing to evaluate the arm-selection objective on 1% of the arms | GLOC-TS · Logit model, N=100,000, T=5000 | cumulative regret | Hurt |
| Use L1 sampling instead of L2 sampling | Approximate inner product estimation · Gaussian query and projection vectors, d=200 | variance: 0.99 to 0.63 | Helped |
| Use L1 sampling instead of L2 sampling | Approximate inner product estimation · 1000-dimensional vectors, m=5 | estimation-error tail thickness | Helped |
Closest papers by meaning
- Generalized Linear Bandits: Almost Optimal Regret with One-Pass Update (NeurIPS 2025 · similarity 0.92)
- Generalized Linear Bandits with Limited Adaptivity (NeurIPS 2024 · similarity 0.89)
- Combinatorial Stochastic-Greedy Bandit (AAAI 2024 · similarity 0.88)
- Heavy-Tailed Linear Bandits: Huber Regression with One-Pass Update (ICML 2025 · similarity 0.88)
- Batched Stochastic Linear Bandits with 1-Bit Communication Constraints (NeurIPS 2026 · similarity 0.87)
- Geometry-Aware Approaches for Balancing Performance and Theoretical Guarantees in Linear Bandits (ICLR 2025 · similarity 0.87)
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.