Unsupervised Non-transferable Text Classification
This work addresses the challenge of protecting intellectual property in neural networks for text classification, offering a solution that is more practical due to its unsupervised nature and recoverability feature.
The paper tackles the problem of preventing unauthorized exploitation of deep learning models by proposing an unsupervised non-transferable learning method for text classification that eliminates the need for labeled target domain data and includes a secret key mechanism for recovering access, achieving effectiveness as demonstrated in experiments.
Training a good deep learning model requires substantial data and computing resources, which makes the resulting neural model a valuable intellectual property. To prevent the neural network from being undesirably exploited, non-transferable learning has been proposed to reduce the model generalization ability in specific target domains. However, existing approaches require labeled data for the target domain which can be difficult to obtain. Furthermore, 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 unsupervised non-transferable learning method for the text classification task that does not require annotated target domain data. We further introduce 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.