Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Augmented RBMLE-UCB Approach for Adaptive Control of Linear Quadratic Systems
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 augmented RBMLE (ball-constrained) instead of unaugmented RBMLE (ablation) | RBMLE · four LQ systems | regret at T=500: 3233 -> 3233; 5930 -> 5930; 16144 -> 16135; 540297 -> 528805 | No clear effect |
| use ARBMLE instead of OFULQ | adaptive LQ control · Unstable Laplacian | regret at T=500: 1.2 \times 10^6 -> 3233 | Helped |
| use ARBMLE instead of Thompson Sampling | adaptive LQ control · UAV | regret at T=500: 1.1 \times 10^20 -> 16135 | Helped |
| use ARBMLE instead of StabL | adaptive LQ control · Boeing 747 | regret at T=500: 1.4 \times 10^{7} -> 528805 | Helped |
| use ARBMLE instead of RCE | adaptive LQ control · UAV | regret at T=500: 180639 -> 16135 | Helped |
| use ARBMLE instead of Input Perturbations (IP) | adaptive LQ control · Boeing 747 | regret at T=500: 540248 -> 528805 | Helped |
| use RBMLE instead of IP on chained integrator | adaptive LQ control · Chained integrator | regret at T=500: 2337 -> 2322 | Helped |
| use RBMLE instead of IP on stabilizable-not-controllable system | adaptive LQ control · Stabilizable not controllable | regret at T=500: 15628 -> 15665 | Hurt |
| use ARBMLE instead of OFULQ on chained integrator | adaptive LQ control · Chained integrator | regret at T=500: 33449 -> 2322 | Helped |
Closest papers by meaning
- Regret Bounds for Robust Adaptive Control of the Linear Quadratic Regulator (NeurIPS 2018 · similarity 0.89)
- Naive Exploration is Optimal for Online LQR (ICML 2020 · similarity 0.88)
- Logarithmic Regret Bound in Partially Observable Linear Dynamical Systems (NeurIPS 2020 · similarity 0.88)
- Regret Bounds for Episodic Risk-Sensitive Linear Quadratic Regulator (ICLR 2025 · similarity 0.88)
- Exploration Through Reward Biasing: Reward-Biased Maximum Likelihood Estimation for Stochastic Multi-Armed Bandits (ICML 2020 · similarity 0.87)
- Noise-Adaptive Confidence Sets for Linear Bandits and Application to Bayesian Optimization (ICML 2024 · similarity 0.85)
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.