Adversarial Reprogramming of Text Classification Neural Networks

Paarth Neekhara, Shehzeen Hussain, Shlomo Dubnov, Farinaz Koushanfar


Abstract
In this work, we develop methods to repurpose text classification neural networks for alternate tasks without modifying the network architecture or parameters. We propose a context based vocabulary remapping method that performs a computationally inexpensive input transformation to reprogram a victim classification model for a new set of sequences. We propose algorithms for training such an input transformation in both white box and black box settings where the adversary may or may not have access to the victim model’s architecture and parameters. We demonstrate the application of our model and the vulnerability of neural networks by adversarially repurposing various text-classification models including LSTM, bi-directional LSTM and CNN for alternate classification tasks.
Anthology ID:
D19-1525
Volume:
Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)
Month:
November
Year:
2019
Address:
Hong Kong, China
Editors:
Kentaro Inui, Jing Jiang, Vincent Ng, Xiaojun Wan
Venues:
EMNLP | IJCNLP
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
5216–5225
Language:
URL:
https://aclanthology.org/D19-1525
DOI:
10.18653/v1/D19-1525
Bibkey:
Cite (ACL):
Paarth Neekhara, Shehzeen Hussain, Shlomo Dubnov, and Farinaz Koushanfar. 2019. Adversarial Reprogramming of Text Classification Neural Networks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 5216–5225, Hong Kong, China. Association for Computational Linguistics.
Cite (Informal):
Adversarial Reprogramming of Text Classification Neural Networks (Neekhara et al., EMNLP-IJCNLP 2019)
Copy Citation:
PDF:
https://aclanthology.org/D19-1525.pdf
Attachment:
 D19-1525.Attachment.pdf
Code
 paarthneekhara/rnn_adversarial_reprogramming
Data
IMDb Movie ReviewsImageNet