Research map / Reinforcement learning / Policy gradient and actor-critic
Robust Reinforcement Learning: A Case Study in Linear Quadratic Regulation
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 |
|---|---|---|---|
| increase rollout count M | O-LSPI · LQR numerical example | fraction of runs with stable policies | Helped |
| increase rollout count M | O-LSPI · LQR numerical example | relative cost error | Helped |
| increase policy-evaluation iterations T | O-LSPI · LQR numerical example | fraction of runs with stable policies | No clear effect |
| increase policy-evaluation iterations T | O-LSPI · LQR numerical example | relative cost error | Helped |
| use O-LSPI instead of LSPIv1 | O-LSPI and LSPIv1 · LQR numerical example, M < 10^4 | relative cost error | Hurt |
| use O-LSPI instead of LSPIv1 | O-LSPI and LSPIv1 · LQR numerical example, M ≥ 10^4 | relative cost error | No clear effect |
| increase exploration variance to σ²u ≥ 10 | O-LSPI and LSPIv1 · LQR numerical example | relative cost error | Helped |
| reduce exploration variance below 10 | O-LSPI and LSPIv1 · LQR numerical example | relative cost error | Hurt |
| use O-LSPI instead of LSPIv1 at very small exploration variance | O-LSPI and LSPIv1 · LQR numerical example, σ²u ≤ 10^-5 | relative cost error | Helped |
Closest papers by meaning
- Robust exploration in linear quadratic reinforcement learning (NeurIPS 2019 · similarity 0.91)
- On the Stability and Convergence of Robust Adversarial Reinforcement Learning: A Case Study on Linear Quadratic Systems (NeurIPS 2020 · similarity 0.90)
- Robust Model Based Reinforcement Learning Using $\mathcal{L}_1$ Adaptive Control (ICLR 2024 · similarity 0.89)
- Certifying Stability of Reinforcement Learning Policies using Generalized Lyapunov Functions (NeurIPS 2025 · similarity 0.89)
- Time-Constrained Robust MDPs (NeurIPS 2024 · similarity 0.88)
- Robust Offline Reinforcement Learning with Linearly Structured $f$-Divergence Regularization (ICML 2025 · similarity 0.88)
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.