الشبكات العصبية العميقة عرضة للهجمات الخصومة، حيث اضطراب صغير في المدخل يغير التنبؤ النموذجي.في كثير من الحالات، يمكن أن تخدع المدخلات الخبيثة عن قصد لنموذج واحد نموذج آخر.في هذه الورقة، نقدم الدراسة الأولى للتحقيق بشكل منهجي في تحويل أمثلة الخصومة بشكل منهجي لنماذج تصنيف النص واستكشاف كيفية تأثير مختلف العوامل، بما في ذلك بنية الشبكة، نظام التكتلات، وإدماج الكلمات، والقدرة النموذجية، على تحويل أمثلة الخصومة.بناء على هذه الدراسات، نقترح خوارزمية وراثية للعثور على مجموعة من النماذج التي يمكن استخدامها لتحفيز أمثلة الخصومة لخداع جميع النماذج الحالية تقريبا.تعكس هذه الأمثلة المخدرة عيوب عملية التعلم وتحيز البيانات في مجموعة التدريب.أخيرا، نحن نستمد قواعد استبدال الكلمات التي يمكن استخدامها لتشخيصات النموذج من هذه الأمثلة الخصومة.
Deep neural networks are vulnerable to adversarial attacks, where a small perturbation to an input alters the model prediction. In many cases, malicious inputs intentionally crafted for one model can fool another model. In this paper, we present the first study to systematically investigate the transferability of adversarial examples for text classification models and explore how various factors, including network architecture, tokenization scheme, word embedding, and model capacity, affect the transferability of adversarial examples. Based on these studies, we propose a genetic algorithm to find an ensemble of models that can be used to induce adversarial examples to fool almost all existing models. Such adversarial examples reflect the defects of the learning process and the data bias in the training set. Finally, we derive word replacement rules that can be used for model diagnostics from these adversarial examples.
References used
https://aclanthology.org/
Recent work has demonstrated the vulnerability of modern text classifiers to universal adversarial attacks, which are input-agnostic sequences of words added to text processed by classifiers. Despite being successful, the word sequences produced in s
We propose the first general-purpose gradient-based adversarial attack against transformer models. Instead of searching for a single adversarial example, we search for a distribution of adversarial examples parameterized by a continuous-valued matrix
Deep learning is at the heart of the current rise of artificial intelligence. In the field of Computer Vision, it has become the workhorse for applications ranging from self-driving cars to surveillance and security. Whereas deep neural networks have
The security of several recently proposed ciphers relies on the fact:" that
the classical methods of cryptanalysis (e.g. linear or differential attacks) are
based on probabilistic characteristics, which makes their security grow
exponentially with
Recent literatures have shown that knowledge graph (KG) learning models are highly vulnerable to adversarial attacks. However, there is still a paucity of vulnerability analyses of cross-lingual entity alignment under adversarial attacks. This paper