RPN: A Word Vector Level Data Augmentation Algorithm in Deep Learning for Language Understanding

TL;DR

The paper introduces RPN, a novel word vector level data augmentation algorithm that enhances natural language understanding models by introducing noise into word embeddings, outperforming existing methods across various tasks.

Contribution

RPN is a new data augmentation technique that operates directly on word vectors, simplifying application to large datasets and improving performance in NLU tasks.

Findings

RPN outperforms existing augmentation methods in multiple NLU tasks.

RPN achieves state-of-the-art results in seven NLU benchmarks.

RPN is effective in low-resource scenarios.

Abstract

Data augmentation is a widely used technique in machine learning to improve model performance. However, existing data augmentation techniques in natural language understanding (NLU) may not fully capture the complexity of natural language variations, and they can be challenging to apply to large datasets. This paper proposes the Random Position Noise (RPN) algorithm, a novel data augmentation technique that operates at the word vector level. RPN modifies the word embeddings of the original text by introducing noise based on the existing values of selected word vectors, allowing for more fine-grained modifications and better capturing natural language variations. Unlike traditional data augmentation methods, RPN does not require gradients in the computational graph during virtual sample updates, making it simpler to apply to large datasets. Experimental results demonstrate that RPN consistently outperforms existing data augmentation techniques across various NLU tasks, including sentiment analysis, natural language inference, and paraphrase detection. Moreover, RPN performs well in low-resource settings and is applicable to any model featuring a word embeddings layer. The proposed RPN algorithm is a promising approach for enhancing NLU performance and addressing the challenges associated with traditional data augmentation techniques in large-scale NLU tasks. Our experimental results demonstrated that the RPN algorithm achieved state-of-the-art performance in all seven NLU tasks, thereby highlighting its effectiveness and potential for real-world NLU applications.