Research map / Reinforcement learning / MDPs and planning
On the Sample Complexity and Metastability of Heavy-tailed Policy Search in Continuous Control
Amrit Singh Bedi, Anjaly Parayil, Junyu Zhang, Mengdi Wang, Alec Koppel
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 |
|---|---|---|---|
| use a Cauchy (heavy-tailed) policy instead of a Gaussian policy | policy gradient · 1D Mario | estimated tail index alpha of gradient: 1 -> 0.2 | Improved |
| use heavy-tailed Cauchy policy (HPG) instead of Gaussian PG | policy gradient · Pathological Mountain Car | escape from spurious local maximum / reward | Improved |
| tune PPO clipping epsilon | PPO · Pathological Mountain Car | cumulative reward | No clear effect |
Closest papers by meaning
- On the Hidden Biases of Policy Mirror Ascent in Continuous Action Spaces (ICML 2022 · similarity 0.92)
- Reinforcement Learning in Continuous Time and Space: A Stochastic Control Approach (JMLR 2020 · similarity 0.90)
- Learning Optimal Deterministic Policies with Stochastic Policy Gradients (ICML 2024 · similarity 0.89)
- REINFORCE Converges to Optimal Policies with Any Learning Rate (NeurIPS 2025 · similarity 0.89)
- Exploring Starts Are Not Enough: Counterexamples and a Fix for Monte Carlo Exploring Starts (NeurIPS 2026 · similarity 0.87)
- Taming "data-hungry" reinforcement learning? Stability in continuous state-action spaces (NeurIPS 2024 · similarity 0.87)
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.
