arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1603.02618cs.CLcs.CV
收藏

红色的!:学习根据事物的区别性来指称事物

The red one!: On learning to refer to things based on their discriminative properties

University of Trento(特伦托大学)

Angeliki Lazaridou, Nghia The Pham, Marco Baroni

摘要:

作为智能体学习交流其视觉环境的第一步,我们提出了一个系统,给定所指对象(猫)和上下文(沙发)的视觉表示,识别其区分属性,即,区分它们的属性(has_tail)。此外,尽管在属性级别缺乏直接监督,但模型学习为对象分配合理的属性(sofa-has_cushion)。最后,我们提出了一个初步的实验,证实了预测的判别属性的参考成功。

英文摘要:

As a first step towards agents learning to communicate about their visual environment, we propose a system that, given visual representations of a referent (cat) and a context (sofa), identifies their discriminative attributes, i.e., properties that distinguish them (has_tail). Moreover, despite the lack of direct supervision at the attribute level, the model learns to assign plausible attributes to objects (sofa-has_cushion). Finally, we present a preliminary experiment confirming the referential success of the predicted discriminative attributes.

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。

↑