Unsupervised Non-transferable Text Classification

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 10071 - 10084 December 7-11, 2022 ©2022 Association for Computational Linguistics Unsupervised Non-transferable Text Classification Guangtao Zeng and Wei Lu StatNLP Research Group Singapore University of Technology and Design guangtao_zeng@mymail.sutd.edu.sg, luwei@sutd.edu.sg Abstract Training a good deep learning model requires substantial data and computing resources, which makes the resulting neural model a valu- able intellectual property. To prevent the neu- ral network from being undesirably exploited, non-transferable learning has been proposed to reduce the model generalization ability in specific target domains. However, existing ap- proaches require labeled data for the target do- main which can be difficult to obtain. Further- more, they do not have the mechanism to still recover the model’s ability to access the target domain. In this paper, we propose a novel unsu- pervised non-transferable learning method for the text classification task that does not require annotated target domain data. We further intro- duce a secret key component in our approach for recovering the access to the target domain, where we design both an explicit and an implicit method for doing so. Extensive experiments demonstrate the effectiveness of our approach. 1 Introduction Deep learning has achieved remarkable success over the past decade and is active in various fields, including computer vision, natural language pro- cessing (NLP), and data mining. Although neural models can perform well in most tasks, they require a huge amount of data and a high computation cost to train, making the trained model a valuable in- tellectual property. As a result, it is essential for us to prevent the neural models from being used without authorization. In the last few years, many methods have been proposed to safeguard deep neural networks and they can be roughly divided into two types: watermarking (Adi et al., 2018), and secure authorization (Alam et al., 2020). In the watermarking approaches, the owners can ver- ify the ownership of the neural model based on a unique watermark. However, due to the catas- trophic forgetting problem (Kemker et al., 2018), the watermark-based neural models (Kuribayashi ������ ������ ��� � ����� ������ ������ ������ ������ ������ ����� ����� ������ Figure 1: Overview of our unsupervised non- transferable with secret keys. et al., 2020; Song et al., 2017) are known to be vulnerable to certain malicious attacks (Wang and Kerschbaum, 2019), which may lead to the loss of their watermarks. On the other hand, in the se- cure authorization approaches, the owners of the neural network want to ensure that users can only access the model with authorization. Recently, Wang et al. (2022) proposed a new perspective with non-transferable learning (NTL) to protect the model from illegal access to unauthorized data. The method trains the model to have a good per- formance only in the authorized domain while per- forming badly in the unauthorized domain. How- ever, such an approach has some limitations: 1) their work relies on a significant amount of labeled data from the target domain, while such …

Content truncated to 500 words