Adversarial Reprogramming of Text Classification Neural Networks

By Paarth Neekhara, Shehzeen Samarah Hussain, Shlomo Dubnov, Farinaz Koushanfar. 2018

49 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Natural Language Processing, Safety, Robustness & Interpretability, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Hrant Khachatrian
2. Jonathan May
3. Karen Hambardzumyan
4. Noah Constant
5. Rami Al-Rfou
6. Brian Lester
7. Phillip Isola
8. S. Sankaranarayanan

In Natural Language Processing

1. Hrant Khachatrian
2. Jonathan May
3. Karen Hambardzumyan

In Safety, Robustness & Interpretability

1. Hrant Khachatrian
2. Jonathan May
3. Karen Hambardzumyan
4. Brian Lester
5. Rami Al-Rfou
6. Noah Constant
7. Zonghan Yang
8. Yang Liu

Citations in the corpus

• WARP: Word-level Adversarial ReProgramming (2021, methodology): “Adversarial reprogramming has been adapted to text classification tasks with LSTM networks in (Neekhara et al., 2019).”