arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 1206.3232cs.AI
收藏

AND/OR重要性采样

AND/OR Importance Sampling

University of California, Irvine(加利福尼亚大学欧文分校)

Vibhav Gogate, Rina Dechter

摘要:

本文介绍了概率图模型的与或重要性采样。与重要性采样相反,与/或重要性采样在与/或空间中缓存样本,然后从存储的样本中提取新的样本均值。我们证明,与/或重要性采样可能有较低的方差比重要性采样,从而提供了一个理论上的理由,更喜欢它超过重要性采样。我们的实证评估表明,与/或重要性采样是更准确的重要性采样在许多情况下。

英文摘要:

The paper introduces AND/OR importance sampling for probabilistic graphical models. In contrast to importance sampling, AND/OR importance sampling caches samples in the AND/OR space and then extracts a new sample mean from the stored samples. We prove that AND/OR importance sampling may have lower variance than importance sampling; thereby providing a theoretical justification for preferring it over importance sampling. Our empirical evaluation demonstrates that AND/OR importance sampling is far more accurate than importance sampling in many cases.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑