Few-shot learning has been used to tackle the problem of label scarcity in text classification, of which meta-learning based methods have shown to be effective, such as the prototypical networks (PROTO). Despite the success of PROTO, there still exist three main problems: (1) ignore the randomness of the sampled support sets when computing prototype vectors; (2) disregard the importance of labeled samples; (3) construct meta-tasks in a purely random manner. In this paper, we propose a Meta-Learning Siamese Network, namely, Meta-SN, to address these issues. Specifically, instead of computing prototype vectors from the sampled support sets, Meta-SN utilizes external knowledge (e.g. class names and descriptive texts) for class labels, which is encoded as the low-dimensional embeddings of prototype vectors. In addition, Meta-SN presents a novel sampling strategy for constructing meta-tasks, which gives higher sampling probabilities to hard-to-classify samples. Extensive experiments are conducted on six benchmark datasets to show the clear superiority of Meta-SN over other state-of-the-art models. For reproducibility, all the datasets and codes are provided at https://github.com/hccngu/Meta-SN.
翻译:在文本分类中,以元学习为基础的方法证明是有效的,例如原型网络(PROTO)。尽管PROTO取得了成功,但仍有三个主要问题:(1) 在计算原型矢量时,忽略抽样支持组的随机性;(2) 无视标签样本的重要性;(3) 以纯随机的方式构建元任务。在本文件中,我们提议建立一个Meta-Learning Siames网络,即Meta-Learn Siames 网络,以解决这些问题。具体地说,Meta-SN(Meta-Learn Siames)不是从抽样支助组中计算原型矢量,而是利用外部知识(如类名称和描述文本)为类标签,该类标签的编码是原型矢量的低维嵌入。此外,Meta-SNS(MSN)为构建元任务的新抽样战略,为难以分类的样品提供更高的采样概率。在六个基准数据集上进行了广泛的实验,以显示Met-SNE(Met-S)相对于其他州-art模型的明显优势。http://Mestrecommissions。