arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1711.00482cs.CLcs.NE
收藏

使用潜在语言学习

Learning with Latent Language

University of California, Berkeley(加州大学伯克利分校)

Jacob Andreas, Dan Klein, Sergey Levine

摘要:

自然语言中的命名概念和组合操作符提供了丰富的信息源,这些信息是关于人类用来导航世界的各种抽象的。这种语言学背景知识能否提高学习分类器和控制策略的通用性和效率?本文旨在表明,使用自然语言串的空间作为参数空间是捕捉自然任务结构的有效途径。在预训练阶段,我们学习一个语言解释模型,该模型将输入(例如图像)转换为给定自然语言描述的输出(例如标签)。为了学习一个新的概念(例如分类器),我们直接在描述空间中搜索,以最小化解释器在训练样本上的损失。至关重要的是,我们的模型不需要语言数据来学习这些概念:语言只在预训练中使用,以便在后续学习中强加结构。图像分类、文本编辑和强化学习的结果表明,在所有设置中,具有语言参数化的模型都优于没有语言参数化的模型。

英文摘要:

The named concepts and compositional operators present in natural language provide a rich source of information about the kinds of abstractions humans use to navigate the world. Can this linguistic background knowledge improve the generality and efficiency of learned classifiers and control policies? This paper aims to show that using the space of natural language strings as a parameter space is an effective way to capture natural task structure. In a pretraining phase, we learn a language interpretation model that transforms inputs (e.g. images) into outputs (e.g. labels) given natural language descriptions. To learn a new concept (e.g. a classifier), we search directly in the space of descriptions to minimize the interpreter's loss on training examples. Crucially, our models do not require language data to learn these concepts: language is used only in pretraining to impose structure on subsequent learning. Results on image classification, text editing, and reinforcement learning show that, in all settings, models with a linguistic parameterization outperform those without.

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。

↑