Abstract
Although attention mechanisms have been applied to a variety of deep learning models and have been shown to improve the prediction performance, it has been reported to be vulnerable to perturbations to the mechanism. To overcome the vulnerability to perturbations in the mechanism, we are inspired by adversarial training (AT), which is a powerful regularization technique for enhancing the robustness of the models. In this paper, we propose a general training technique for natural language processing tasks, including AT for attention (Attention AT) and more interpretable AT for attention (Attention iAT). The proposed techniques improved the prediction performance and the model interpretability by exploiting the mechanisms with AT. In particular, Attention iAT boosts those advantages by introducing adversarial perturbation, which enhances the difference in the attention of the sentences. Evaluation experiments with ten open datasets revealed that AT for attention mechanisms, especially Attention iAT, demonstrated (1) the best performance in nine out of ten tasks and (2) more interpretable attention (i.e., the resulting attention correlated more strongly with gradient-based word importance) for all tasks. Additionally, the proposed techniques are (3) much less dependent on perturbation size in AT. Our code is available at https://github.com/shunk031/attention-meets-perturbation
Highlights
A TTENTION mechanisms [1] are widely applied in natural language processing (NLP) field through deep neural networks (DNNs)
COMMON MODEL ARCHITECTURE Our goal is to improve the performance of NLP models by aiming at the robustness of the attention mechanisms
For 20 Newsgroups (20News) and AG News (AGNews) in the binary classification (BC) and bAbI task 1 in question answering (QA), the conventional techniques, including the Vanilla model, were sufficiently accurate, so the performance improvement of the proposed techniques to the tasks was limited to some extent
Summary
A TTENTION mechanisms [1] are widely applied in natural language processing (NLP) field through deep neural networks (DNNs). The Transformer [8] and its follow-up models [12], [13] have self-attention mechanisms that estimate the relationship of each word in the sentence. These models take advantage of the effect of the mechanisms and have shown promising performances. There is no doubt that the effect of the mechanisms is extremely large They are not easy to train, as they require huge amounts of GPU memory to maintain the weights of the model. The application of attention mechanisms to DNN models, such as RNN and CNN models, which have been widely used and do not require relatively high training requirements, has not been sufficiently studied
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
