Research map / Optimization / Nonconvex and smooth optimization
Bregman Gradient Policy Optimization
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 diagonal mirror mapping instead of l_p-norm mapping (ablation) | BGPO · CartPole-v1, Acrobot-v1, MountainCarContinuous-v0 | average return | Helped |
| Use l_p-norm mirror mapping, varied over p = (1.50, 2.0, 3.0), instead of diagonal mapping (ablation) | BGPO · CartPole-v1, Acrobot-v1, MountainCarContinuous-v0 | average return | Hurt |
| Add the variance-reduced gradient estimator (ablation) | BGPO · CartPole-v1 | average return | No clear effect |
| Add the variance-reduced gradient estimator (ablation) | BGPO · Acrobot-v1 and MountainCarContinuous-v0 | average return | Helped |
| Add the variance-reduced gradient estimator (ablation) | BGPO · CartPole-v1, Acrobot-v1, MountainCarContinuous-v0 | training stability | Helped |
| Use VR-BGPO with diagonal mapping instead of the evaluated baseline methods | VR-BGPO · Six continuous-control Gym environments | average return | Helped |
| Use BGPO with diagonal mapping instead of the evaluated baseline methods | BGPO · Six continuous-control Gym environments | average return | Mixed |
| Use l_p-norm mapping for the mirror function | VRMPO · Six continuous-control Gym environments | average return | Mixed |
Closest papers by meaning
- Momentum-Based Policy Gradient Methods (ICML 2020 · similarity 0.89)
- Natural Policy Gradient as Doubly Smoothed Policy Iteration: A Bellman-Operator Framework (NeurIPS 2026 · similarity 0.87)
- Accelerated Policy Gradient: On the Convergence Rates of the Nesterov Momentum for Reinforcement Learning (ICML 2024 · similarity 0.87)
- On the Second-Order Convergence of Biased Policy Gradient Algorithms (ICML 2024 · similarity 0.86)
- Convergence Analysis of Policy Gradient Methods with Dynamic Stochasticity (ICML 2025 · similarity 0.86)
- Diffusion Policy Policy Optimization (ICLR 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.