Augmenting Data with Mixup for Sentence Classification: An Empirical Study
This addresses the problem of data augmentation for NLP tasks, specifically sentence classification, but is incremental as it adapts an existing method from computer vision.
The paper tackled applying Mixup data augmentation to sentence classification by proposing two interpolation strategies on word and sentence embeddings, resulting in significant accuracy improvements for CNN and LSTM models on benchmark datasets.
Mixup, a recent proposed data augmentation method through linearly interpolating inputs and modeling targets of random samples, has demonstrated its capability of significantly improving the predictive accuracy of the state-of-the-art networks for image classification. However, how this technique can be applied to and what is its effectiveness on natural language processing (NLP) tasks have not been investigated. In this paper, we propose two strategies for the adaption of Mixup on sentence classification: one performs interpolation on word embeddings and another on sentence embeddings. We conduct experiments to evaluate our methods using several benchmark datasets. Our studies show that such interpolation strategies serve as an effective, domain independent data augmentation approach for sentence classification, and can result in significant accuracy improvement for both CNN and LSTM models.