Learning from limited datasets: Implications for Natural Language Generation and Human-Robot Interaction
Learning from limited datasets: Implications for Natural Language Generation and Human-Robot Interaction
复制标题
从有限的数据集中学习：对自然语言生成和人机交互的影响
DOI：
--
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
Dimitra Gkatzia
中科院分区：
文献类型：
--
作者：
Jekaterina Belakova;Dimitra Gkatzia
人类机器人最自然的交流方式之一是通过口语。训练人-机器人交互系统需要访问大型数据集，这些数据集的获取成本很高，而且是劳动密集型的。在本文中，我们描述了一种从最小数据学习的方法，并以口语对话系统中的语言理解为例。理解口语是至关重要的，因为它对自然语言生成有影响，即正确理解用户的话语将导致选择正确的反应/动作。最后，我们讨论了人-机器人交互对自然语言生成的影响。
One of the most natural ways for human robot communication is through spoken language. Training human-robot interaction systems require access to large datasets which are expensive to obtain and labour intensive. In this paper, we describe an approach for learning from minimal data, using as a toy example language understanding in spoken dialogue systems. Understanding of spoken language is crucial because it has implications for natural language generation, i.e. correctly understanding a user’s utterance will lead to choosing the right response/action. Finally, we discuss implications for Natural Language Generation in Human-Robot Interaction.