Research map / Reinforcement learning / Policy gradient and actor-critic
Adaptive Smoothing for Path Integral Control
Dominik Thalmeier, Hilbert J. Kappen, Simone Totaro, Vicenç Gómez
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 |
|---|---|---|---|
| add cost smoothing (ASPIC) to direct cost optimization (ablation) | policy optimization with trust region · Pendulum swing-up, Acrobot, 2D walker | iterations to reach a given cost | Improved |
| use intermediate smoothing strength instead of none or very strong (ablation) | ASPIC · LQ Brownian viapoints, threshold C<=2*10^4 | iterations to threshold | Mixed |
| reduce rollouts per iteration from N=500 to N=50 under fixed 25K budget (ablation) | ASPIC · Pendulum swing-up | final cost: >350 -> < -50 | Improved |
| increase trust region size E (ablation) | ASPIC · Acrobot (N=500) | convergence speed | Improved |
| reduce smoothing strength Delta (ablation) | ASPIC · Acrobot (N=500) | convergence stability | Hurt |
Closest papers by meaning
- Memory-Augmented Potential Field Theory: A Framework for Adaptive Control in Non-Convex Domains (NeurIPS 2025 · similarity 0.84)
- ART for Diffusion Sampling: A Reinforcement Learning Approach to Timestep Scheduling (NeurIPS 2026 · similarity 0.84)
- DiffPhyCon: A Generative Approach to Control Complex Physical Systems (NeurIPS 2024 · similarity 0.84)
- Impact of Computation in Integral Reinforcement Learning for Continuous-Time Control (ICLR 2024 · similarity 0.83)
- Score-Based Diffusion Policy Compatible with Reinforcement Learning via Optimal Transport (ICML 2025 · similarity 0.83)
- PDPO: Parametric Density Path Optimization (NeurIPS 2025 · similarity 0.83)
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.
