Are we Bayesian referring expression generators
Are we Bayesian referring expression generators
复制标题
我们是贝叶斯引用表达式生成器吗
DOI：
--
复制
发表时间：
2013
期刊：
影响因子：
--
通讯作者：
E. Krahmer
中科院分区：
文献类型：
--
作者：
Albert Gatt;R. V. Gompel;Kees van Deemter;E. Krahmer
Frank和Goodman(2012)最近的一篇论文提出了一个简单参照游戏的贝叶斯模型。该模型中包含的权利要求之一是，选择使用哪个词或属性来指代对象取决于该属性的效用。在本文中，我们将该模型与参考生产的其他计算模型，特别是最近的PRO(概率参考覆盖)模型进行了比较。我们认为，弗兰克和古德曼(2012)模型中指导物业选择的效用假设是不充分的，因为它忽略了过度投机的可能性和物业之间偏好排名的作用，因此，无论它们的效用如何，它们都可能被使用。我们表明，确实考虑到这一点的模型，如PRO，对参与者有可能过度指定的实验数据有更好的检验。
A recent paper by Frank and Goodman (2012) proposes a Bayesian model of simple referential games. One of the claims embodied in the model is that choosing which word or property to use to refer to an object depends on the utility of the property. In this paper, we compare this model to other computational models of reference production, in particular the recent pro (Probabilistic Referential Overspecication) model. We argue that the assumption of utility that guides property choice in the Frank and Goodman (2012) model is inadequate, insofar as it ignores the possibility of overspecication and the role of preference rankings among properties, as a result of which they may be used irrespective of their utility. We show that models that do take this into account, such as pro, have a better t to experimental data in which participants have the possibility of overspecifying.