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 1906.00451cs.LGstat.ML
收藏

结构化预测中的精确推理

Exact inference in structured prediction

Kevin Bello, Jean Honorio

摘要:

结构化预测可以被认为是多个标签的同时预测。这通常通过最大化标签空间上的得分函数来完成,该得分函数分解为成对和一元势的总和。以上自然地用图来建模,其中边和顶点分别与成对势和一元势相关。我们考虑了Globerson等人提出的生成过程,并将其应用于一般连通图。我们分析的结构条件的图,允许的标签的准确恢复。我们的研究结果表明,准确的恢复是可能的,并在多项式时间的一个大类的图实现。特别是,我们表明,图是坏的扩展器可以通过添加来自Erdens-Rényi模型的小的边缘扰动来完全恢复。最后,作为我们分析的副产品,我们提供了一个扩展的Cheeger不等式。

英文摘要:

Structured prediction can be thought of as a simultaneous prediction of multiple labels. This is often done by maximizing a score function on the space of labels, which decomposes as a sum of pairwise and unary potentials. The above is naturally modeled with a graph, where edges and vertices are related to pairwise and unary potentials, respectively. We consider the generative process proposed by Globerson et al. and apply it to general connected graphs. We analyze the structural conditions of the graph that allow for the exact recovery of the labels. Our results show that exact recovery is possible and achievable in polynomial time for a large class of graphs. In particular, we show that graphs that are bad expanders can be exactly recovered by adding small edge perturbations coming from the Erdős-Rényi model. Finally, as a byproduct of our analysis, we provide an extension of Cheeger's inequality.

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。

↑