PulseAugur
实时 15:05:21

Score-Repellent Monte Carlo 提供了高效的、具有恒定内存的非马尔可夫采样

研究人员推出了一种名为 Score-Repellent Monte Carlo (SRMC) 的新颖框架,旨在提高通用状态空间中非马尔可夫采样的效率。SRMC 通过对评分评估进行运行平均来总结轨迹历史,从而能够阻止冗余的重访并降低蒙特卡洛方差。与现有方法相比,该方法在保持恒定内存使用的同时,在连续目标和离散能量模型上的实验证明了其改进的估计量方差和模式覆盖率。 AI

影响 引入了一种新的采样技术,可以提高 AI 模型训练和推理的效率并降低方差。

排序理由 这是一篇详细介绍蒙特卡洛采样新算法框架的研究论文。

在 arXiv stat.ML 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

Score-Repellent Monte Carlo 提供了高效的、具有恒定内存的非马尔可夫采样

报道来源 [2]

  1. arXiv stat.ML TIER_1 English(EN) · Jie Hu, Lingyun Chen, Geeho Kim, Jinyoung Choi, Bohyung Han, Do Young Eun ·

    Score-Repellent Monte Carlo: Toward Efficient Non-Markovian Sampler with Constant Memory in General State Spaces

    arXiv:2604.22948v1 Announce Type: cross Abstract: History-dependent sampling can reduce long-run Monte Carlo variance by discouraging redundant revisits, but existing schemes typically encode history through empirical measure on finite state spaces, which is infeasible in high-di…

  2. arXiv stat.ML TIER_1 English(EN) · Do Young Eun ·

    Score-Repellent Monte Carlo: Toward Efficient Non-Markovian Sampler with Constant Memory in General State Spaces

    History-dependent sampling can reduce long-run Monte Carlo variance by discouraging redundant revisits, but existing schemes typically encode history through empirical measure on finite state spaces, which is infeasible in high-dimensional discrete configuration spaces or ill-pos…