Proactive Detection of Query-based Adversarial Scenarios in NLP Systems

Mohammad Maghsoudimehrabani, Amin Azmoodeh, Ali Dehghantanha, Behrouz Zolfaghari, Gautam Srivastava

研究成果: Conference contribution

2 被引用数 (Scopus)

抄録

Adversarial attacks can mislead a Deep Learning (DL) algorithm into generating erroneous predictions via feeding maliciously-disturbed inputs called adversarial examples. DL-based Natural Language Processing (NLP) algorithms are severely threatened by adversarial attacks. In real-world, black-box adversarial attacks, the adversary needs to submit many highly-similar queries before drafting an adversarial example. Due to this long process, in-progress attack detection can play a significant role in adversarial defense in DL-based NLP algorithms. Although there are several approaches for detecting adversarial attacks in NLP, these approaches are reactive in the sense that they can detect adversarial examples only when they are fabricated and fed into the algorithm. In this study, we take one step towards proactive detection of adversarial attacks in NLP systems by proposing a robust, history-based model named Stateful Query Analysis (SQA) to identify suspiciously-similar sequences of queries capable of generating textual adversarial examples to which we refer by adversarial scenarios. The model exhibits a detection rate of over 99.9% in our extensive experimental tests against several state-of-The-Art black-box adversarial attack methods.

本文言語English
ホスト出版物のタイトルAISec 2022 - Proceedings of the 15th ACM Workshop on Artificial Intelligence and Security, co-located with CCS 2022
出版社Association for Computing Machinery, Inc
ページ103-113
ページ数11
ISBN(電子版)9781450398800
DOI
出版ステータスPublished - 2022 11月 11
イベント15th ACM Workshop on Artificial Intelligence and Security, AISec 2022 - Co-located with CCS 2022 - Los Angeles, United States
継続期間: 2022 11月 11 → …

出版物シリーズ

名前AISec 2022 - Proceedings of the 15th ACM Workshop on Artificial Intelligence and Security, co-located with CCS 2022

Conference

Conference15th ACM Workshop on Artificial Intelligence and Security, AISec 2022 - Co-located with CCS 2022
国/地域United States
CityLos Angeles
Period22/11/11 → …

ASJC Scopus subject areas

  • 人工知能
  • コンピュータ ネットワークおよび通信
  • ソフトウェア

フィンガープリント

「Proactive Detection of Query-based Adversarial Scenarios in NLP Systems」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル