arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 0712.1402cs.CCcs.LG
收藏

从样本重构马尔可夫随机场:一些简单的观察和算法

Reconstruction of Markov Random Fields from Samples: Some Easy Observations and Algorithms

U.C. Berkeley(加州大学伯克利分校)

Guy Bresler, Elchanan Mossel, Allan Sly

摘要:

马尔可夫随机场在许多应用领域被用来模拟高维分布。最近的兴趣一直致力于从马尔可夫随机场的独立样本的依赖结构的重建。我们分析了一个简单的算法,用于重建的底层图形定义的马尔可夫随机场上$n$节点和最大程度的$d$给定的意见。我们表明,在温和的非退化条件下,它重建生成图与高概率使用$Θ(d ε^{-2}δ^{-4} \log n)$样本,其中$ε,δ$依赖于当地的相互作用。对于大多数局部相互作用,$\eps,δ$的顺序为$\exp(-O(d))$。我们的结果是最佳的作为$n$的函数,取决于$d$和本地相互作用的强度的乘法常数。我们的结果似乎是第一个结果,一般的模型,保证{ \em的}生成模型重建。此外,我们提供了明确的$O(n^{d+2} ε^{-2}δ^{-4} \log n)$运行时间限制。在图上的度量具有相关性衰减的情况下,对于所有固定的$d$,运行时间为$O(n^2 \log n)$。我们还讨论了观察噪声样本的效果,并表明,只要噪声水平低,我们的算法是有效的。另一方面,我们构造了一个例子,其中大噪声意味着不可识别性,即使是通用的噪声和相互作用。最后,我们简要地表明,在一些简单的情况下,具有隐藏节点的模型也可以恢复。

英文摘要:

Markov random fields are used to model high dimensional distributions in a number of applied areas. Much recent interest has been devoted to the reconstruction of the dependency structure from independent samples from the Markov random fields. We analyze a simple algorithm for reconstructing the underlying graph defining a Markov random field on $n$ nodes and maximum degree $d$ given observations. We show that under mild non-degeneracy conditions it reconstructs the generating graph with high probability using $Θ(d ε^{-2}δ^{-4} \log n)$ samples where $ε,δ$ depend on the local interactions. For most local interaction $\eps,δ$ are of order $\exp(-O(d))$. Our results are optimal as a function of $n$ up to a multiplicative constant depending on $d$ and the strength of the local interactions. Our results seem to be the first results for general models that guarantee that {\em the} generating model is reconstructed. Furthermore, we provide explicit $O(n^{d+2} ε^{-2}δ^{-4} \log n)$ running time bound. In cases where the measure on the graph has correlation decay, the running time is $O(n^2 \log n)$ for all fixed $d$. We also discuss the effect of observing noisy samples and show that as long as the noise level is low, our algorithm is effective. On the other hand, we construct an example where large noise implies non-identifiability even for generic noise and interactions. Finally, we briefly show that in some simple cases, models with hidden nodes can also be recovered.

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。

↑