Research map / Reinforcement learning / Policy gradient and actor-critic
Marginal Policy Gradients: A Unified Family of Estimators for Bounded Action Spaces with Applications
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 |
|---|---|---|---|
| Use the angular policy gradient instead of the standard multivariate Gaussian policy gradient | A2C · Platform2D-v1 | cumulative discounted reward | Helped |
| Use the angular policy gradient instead of the standard policy gradient | A3C · King of Glory 1v1 | win percentage | Helped |
| Use the angular policy gradient instead of the standard policy gradient | A3C · King of Glory 1v1 | cumulative discounted reward | Helped |
| Replace the sampling-distribution score with the marginal angular score | Angular policy gradient · Platform2D-v1 and King of Glory 1v1 | gradient-estimator variance: roughly 1/2 | Helped |
| Use the angular Gaussian policy instead of the bivariate Gaussian policy | A2C · Platform2D-v1 | convergence speed | Helped |
| Use the angular Gaussian policy instead of the Gaussian policy over the angle | A2C · Platform2D-v1 | cumulative discounted reward | Helped |
Closest papers by meaning
- Expected Policy Gradients (AAAI 2018 · similarity 0.89)
- Accelerated Policy Gradient: On the Convergence Rates of the Nesterov Momentum for Reinforcement Learning (ICML 2024 · similarity 0.89)
- Fourier Policy Gradients (ICML 2018 · similarity 0.89)
- Variance Reduction for Policy Gradient with Action-Dependent Factorized Baselines (ICLR 2018 · similarity 0.89)
- A Theoretical Understanding of Gradient Bias in Meta-Reinforcement Learning (NeurIPS 2022 · similarity 0.89)
- Deep Policy Gradient Methods Without Batch Updates, Target Networks, or Replay Buffers (NeurIPS 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.