Research map / Reinforcement learning / MDPs and planning
Data-Efficient Reinforcement Learning in Continuous State-Action Gaussian-POMDPs
Rowan McAllister, Carl Edward Rasmussen
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 |
|---|---|---|---|
| Plan in belief space with filtered multi-step prediction and filtered execution | PILCO · Cartpole swing-up, common dataset | empirical cost: min z = 3.51 | Improved |
| Filter only during execution (Deisenroth and Peters) vs no filtering (ablation) | PILCO · Cartpole swing-up, common dataset | empirical cost | Improved |
| Add filtered multi-step prediction on top of filtered execution (ablation) | Deisenroth and Peters · Cartpole swing-up, common dataset | empirical cost: gain -39.1 -> -81.7 N/rad | Improved |
| Use MAP-trajectory prediction ignoring model uncertainty (Dallaire) (ablation) | filtered PILCO · Cartpole swing-up, common dataset | predicted vs empirical cost | Hurt |
| Let each algorithm collect its own data (15 trials, 10 seeds) | Filtered PILCO · Cartpole swing-up full RL | empirical loss | Improved |
| Dallaire method with self-collected data | Dallaire et al. · Cartpole swing-up full RL | predicted loss | Hurt |
| Increase observation noise factor k from 1 | unfiltered PILCO · Cartpole swing-up noise sweep | empirical loss | Hurt |
| Low noise factor k = 1, filtering vs none (ablation) | Filtered PILCO · Cartpole swing-up noise sweep | empirical loss | No clear effect |
| Increase observation noise factor k | Filtered PILCO · Cartpole swing-up noise sweep | empirical loss | No clear effect |
Closest papers by meaning
- Sequential Monte Carlo for Policy Optimization in Continuous POMDPs (NeurIPS 2025 · similarity 0.89)
- Provable Representation with Efficient Planning for Partially Observable Reinforcement Learning (ICML 2024 · similarity 0.89)
- Scalable Policy-Based RL Algorithms for POMDPs (NeurIPS 2025 · similarity 0.88)
- How to Explore with Belief: State Entropy Maximization in POMDPs (ICML 2024 · similarity 0.88)
- Sample-Efficient Learning of POMDPs with Multiple Observations In Hindsight (ICLR 2024 · similarity 0.88)
- Model-based Reinforcement Learning for Confounded POMDPs (ICML 2024 · 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.
