Research map / Reinforcement learning / Visual and representation RL
Active Observing in Continuous-time Control
Samuel Holt, Alihan Hüyük, Mihaela van der Schaar
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 |
|---|---|---|---|
| replace Discrete Monitoring with continuous-time Active Observing Control (ablation) | observing policy · Cancer | normalized utility U: 91±0.532 -> 105±0.18 | Improved |
| replace Discrete Monitoring with continuous-time Active Observing Control (ablation) | observing policy · Pendulum | normalized utility U: 130±2.52 -> 177±2.18 | Improved |
| replace Discrete Monitoring with continuous-time Active Observing Control (ablation) | observing policy · Cartpole | normalized reward R: 82.9±0.532 -> 99.5±0.774 | Improved |
| replace Discrete Monitoring with continuous-time Active Observing Control (ablation) | observing policy · Cancer | normalized reward R: 85.7 ± 0.526 -> 98.8 ± 0.174 | Improved |
| replace regular observing (Continuous Planning) with uncertainty-triggered observing (AOC) (ablation) | observing policy · Cancer | observations taken O: 13±0 -> 3.37±0.0302 | Improved |
| replace regular observing (Continuous Planning) with AOC (ablation) | observing policy · Pendulum | normalized utility U: 100±0.904 -> 177±2.18 | Improved |
| replace regular observing with AOC (ablation) | observing policy · Acrobot | normalized reward R: 100±0.462 -> 90.8±0.878 | Hurt |
| replace Discrete Planning with Discrete Monitoring (uncertainty threshold) (ablation) | observing policy · Cancer | normalized reward R: 91.7±0.368 -> 85.8±0.522 | Hurt |
| replace Discrete Planning with Discrete Monitoring (ablation) | observing policy · Pendulum | normalized utility U: 87.2±0.962 -> 130±2.52 | Improved |
| Continuous Planning forced to take 3 observations at regular times instead of AOC's placement (ablation) | Continuous Planning · Cancer | normalized reward R: 98.8±0.173 -> 95.6±0.234 | Hurt |
| Continuous Planning forced to take 4 observations at regular times instead of AOC's placement (ablation) | Continuous Planning · Cancer | normalized reward R: 98.8±0.173 -> 97.3±0.226 | Hurt |
| Continuous Planning with 3 observations vs AOC on utility (ablation) | Continuous Planning · Cancer | normalized utility U: 105±0.183 -> 102±0.234 | Hurt |
Closest papers by meaning
- Finite Sample Analyses for Continuous-time Linear Systems: System Identification and Online Control (NeurIPS 2025 · similarity 0.85)
- Robust Active Measuring under Model Uncertainty (AAAI 2024 · similarity 0.84)
- When to Sense and Control? A Time-adaptive Approach for Continuous-Time RL (NeurIPS 2024 · similarity 0.83)
- Amortized Control of Continuous State Space Feynman-Kac Model for Irregular Time Series (ICLR 2025 · similarity 0.83)
- Time-Aware World Model for Adaptive Prediction and Control (ICML 2025 · similarity 0.82)
- Sequential Monte Carlo for Policy Optimization in Continuous POMDPs (NeurIPS 2025 · similarity 0.82)
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.
