arXiv:2510.03679cs.LGstat.ML2025-10被引 1

GPG无需价值网络,用分组蒙特卡洛估计优势,提升强化学习效率。

Group Policy Gradient

  • 用分组蒙特卡洛估计替代价值函数,免去训练批评家的开销。
  • 在标准基准上表现优于或媲美PPO,且计算资源利用率更高。
  • 适合追求高效训练、减少模型复杂度的研究者与工程应用。

我们提出群策略梯度(Group Policy Gradient, GPG),一种适用于一般马尔可夫决策过程的无批评家策略梯度估计器。受基于人类反馈强化学习(RLHF)中GRPO方法成功的启发,GPG以基于分组的蒙特卡洛优势估计替代了学习的价值函数,消除了训练批评家所带来的内存、计算和超参数调优成本,同时保留了PPO的截断目标结构。我们证明了GPG估计器的一致性,分析了偏差-方差权衡,并通过实验证明GPG在标准基准上表现匹配或超越PPO。GPG更高效利用并行模拟,结合无批评家设计,整体计算资源使用效率高于PPO。

原文摘要 · Abstract (English)

We introduce Group Policy Gradient (GPG), a family of critic-free policy-gradient estimators for general MDPs. Inspired by the success of GRPO's approach in Reinforcement Learning from Human Feedback (RLHF), GPG replaces a learned value function with a group-based Monte Carlo advantage estimator, removing the memory, compute, and hyperparameter costs of training a critic while preserving PPO's clipped-objective structure. We prove the consistency of the GPG estimator, analyze the bias-variance tradeoffs, and demonstrate empirically that GPG matches or outperforms PPO on standard benchmarks. GPG makes better use of parallel simulations, which, together with its critic-free design, results in more efficient use of computational resources than PPO.

强化学习策略梯度无批评家高效训练

Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。