Research map / Reinforcement learning / Policy gradient and actor-critic
An operator view of policy gradient methods
Dibya Ghosh, Marlos C. Machado, Nicolas Le Roux
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 |
|---|---|---|---|
| Pair polynomial-return improvement with KL projection instead of the matching α-divergence projection (ablation) | Op-Reinforce with polynomial returns · Four-room domain | learning speed | Improved |
| Pair polynomial-return improvement with KL projection (ablation) | Op-Reinforce with polynomial returns · Four-room domain | convergence quality | Hurt |
| Use the corresponding α-divergence projection with polynomial-return improvement (ablation) | Op-Reinforce with polynomial returns · Four-room domain | learning speed | Improved |
| Use the corresponding α-divergence projection with polynomial-return improvement (ablation) | Op-Reinforce with polynomial returns · Four-room domain | convergence quality | Improved |
| Anneal α to 1 during optimization while using polynomial-return improvement (ablation) | Op-Reinforce with polynomial returns · Four-room domain | learning speed | Improved |
| Anneal α to 1 during optimization while using polynomial-return improvement (ablation) | Op-Reinforce with polynomial returns · Four-room domain | convergence quality | Improved |
| Increase return variance with entropy regularization | Policy-gradient methods · Policy-gradient learning | improvement per update | Improved |
| Use an almost-deterministic policy in a deterministic environment | Policy-gradient methods · Deterministic policy and environment | learning progress | Hurt |
Closest papers by meaning
- On the Theory of Policy Gradient Methods: Optimality, Approximation, and Distribution Shift (JMLR 2021 · similarity 0.91)
- A Parametric Class of Approximate Gradient Updates for Policy Optimization (ICML 2022 · similarity 0.90)
- Natural Policy Gradient as Doubly Smoothed Policy Iteration: A Bellman-Operator Framework (NeurIPS 2026 · similarity 0.88)
- On the Global Optimality of Policy Gradient Methods in General Utility Reinforcement Learning (NeurIPS 2025 · similarity 0.87)
- Convergence of Policy Mirror Descent Beyond Compatible Function Approximation (ICML 2025 · similarity 0.86)
- Convergence Analysis of Policy Gradient Methods with Dynamic Stochasticity (ICML 2025 · similarity 0.86)
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.
