arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1207.2328cs.LGcond-mat.stat-mechphysics.data-anstat.ML
收藏

随机区组模型中隐类推理的比较研究

Comparative Study for Inference of Hidden Classes in Stochastic Block Models

CNRS(法国国家科学研究中心);ESPCI ParisTech(巴黎高等物理化工学院);University of Würzburg(维尔茨堡大学);CEA Saclay(法国原子能和替代能源委员会萨克雷中心);Institut de Physique Théorique, IPhT(理论物理研究所(IPhT))

Pan Zhang, Florent Krzakala, Jörg Reichardt, Lenka Zdeborová

摘要:

随机块模型中隐类的推断是一个经典问题,有着重要的应用。最常用的方法,这个问题涉及天真的平均场方法或启发式谱方法。最近,置信传播被提出来解决这个问题。在这篇文章中,我们对这三种方法在合成网络上进行了比较研究。我们发现,信念传播表现出更好的性能相比,天真的平均场和频谱的方法。这适用于准确性,计算效率和过度拟合数据的趋势。

英文摘要:

Inference of hidden classes in stochastic block model is a classical problem with important applications. Most commonly used methods for this problem involve na\"ıve mean field approaches or heuristic spectral methods. Recently, belief propagation was proposed for this problem. In this contribution we perform a comparative study between the three methods on synthetically created networks. We show that belief propagation shows much better performance when compared to na\"ıve mean field and spectral approaches. This applies to accuracy, computational efficiency and the tendency to overfit the data.

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。

↑