Research map / Reinforcement learning / Policy gradient and actor-critic
On Stationary Point Convergence of PPO-Clip
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Theoretical analysis proving stationary point convergence of the PPO-Clip algorithm in reinforcement learning.
Problem
The theoretical understanding of PPO-Clip, the most frequently implemented PPO variant with its clipped surrogate objective, is limited, particularly regarding convergence.
Approach
Provides a convergence analysis of PPO-Clip to stationary points, addressing the clipped surrogate objective that is not typical in other RL algorithms.
Experimental setup
Builds on: proximal policy optimization
Closest papers by meaning
- PPO in the Fisher-Rao geometry (NeurIPS 2026 · similarity 0.91)
- PPO-Clip Attains Global Optimality: Towards Deeper Understandings of Clipping (AAAI 2024 · similarity 0.91)
- Simple Policy Optimization (ICML 2025 · similarity 0.87)
- Convergence Analysis of Policy Gradient Methods with Dynamic Stochasticity (ICML 2025 · similarity 0.87)
- DPO Meets PPO: Reinforced Token Optimization for RLHF (ICML 2025 · similarity 0.86)
- On-Policy Algorithms for Continual Reinforcement Learning (Student Abstract) (AAAI 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.