Structured Output Learning with High Order Loss Functions
Structured Output Learning with High Order Loss Functions
复制标题
具有高阶损失函数的结构化输出学习
DOI：
--
复制
发表时间：
2012
期刊：
影响因子：
--
通讯作者：
R. Zemel
中科院分区：
文献类型：
--
作者：
Daniel Tarlow;R. Zemel
在对结构化域进行建模时，通常需要利用那些不自然地表示为简单标签的信息。示例包括关于将在测试时使用的评估度量的知识，以及部分（弱）标签信息。当附加信息具有根据变量的小子集进行因子分解的结构时（即，是低阶的或可分解的），可以使用几种方法将其并入学习过程中。我们在这项工作中的重点是更具挑战性的情况下，额外的信息不分解根据低阶图形模型结构，我们称之为高阶情况。我们建议将这些额外信息的各种形式化为高阶损失函数，这些损失函数可能在大的变量子集上具有复杂的相互作用。然后，我们解决了计算的挑战，学习中固有的根据这样的损失函数，特别是集中在损失增强的推理问题，出现在大利润率的学习，我们表明，学习高阶损失函数往往是实用的，给出强有力的实证结果，一个流行的和几个新的高阶损失函数，在几个设置。
Often when modeling structured domains, it is desirable to leverage information that is not naturally expressed as simply a label. Examples include knowledge about the evaluation measure that will be used at test time, and partial (weak) label information. When the additional information has structure that factorizes according to small subsets of variables (i.e., is low order, or decomposable), several approaches can be used to incorporate it into a learning procedure. Our focus in this work is the more challenging case, where the additional information does not factorize according to low order graphical model structure; we call this the high order case. We propose to formalize various forms of this additional information as high order loss functions, which may have complex interactions over large subsets of variables. We then address the computational challenges inherent in learning according to such loss functions, particularly focusing on the loss-augmented inference problem that arises in large margin learning; we show that learning with high order loss functions is often practical, giving strong empirical results, with one popular and several novel high-order loss functions, in several settings.