Seeing Is Not Always Believing: Invisible Collision Attack and Defence on Pre-Trained Models

• URL: http://arxiv.org/abs/2309.13579v2
• Date: Tue, 7 May 2024 16:27:05 GMT
• Title: Seeing Is Not Always Believing: Invisible Collision Attack and Defence on Pre-Trained Models
• Authors: Minghang Deng, Zhong Zhang, Junming Shao,
• Abstract summary: Existing backdoor attacks or data poisoning methods often build up the assumption that the attacker invades the computers of victims or accesses the target data. In this paper, we propose a novel framework for an invisible attack on PTMs with enhanced MD5 collision. We extensively validate the effectiveness and stealthiness of our proposed attack and defensive method on different models and data sets.
• Score: 7.7318705389136655
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large-scale pre-trained models (PTMs) such as BERT and GPT have achieved great success in diverse fields. The typical paradigm is to pre-train a big deep learning model on large-scale data sets, and then fine-tune the model on small task-specific data sets for downstream tasks. Although PTMs have rapidly progressed with wide real-world applications, they also pose significant risks of potential attacks. Existing backdoor attacks or data poisoning methods often build up the assumption that the attacker invades the computers of victims or accesses the target data, which is challenging in real-world scenarios. In this paper, we propose a novel framework for an invisible attack on PTMs with enhanced MD5 collision. The key idea is to generate two equal-size models with the same MD5 checksum by leveraging the MD5 chosen-prefix collision. Afterwards, the two ``same" models will be deployed on public websites to induce victims to download the poisoned model. Unlike conventional attacks on deep learning models, this new attack is flexible, covert, and model-independent. Additionally, we propose a simple defensive strategy for recognizing the MD5 chosen-prefix collision and provide a theoretical justification for its feasibility. We extensively validate the effectiveness and stealthiness of our proposed attack and defensive method on different models and data sets.

Related papers

• Privacy Backdoors: Enhancing Membership Inference through Poisoning Pre-trained Models [112.48136829374741]
  In this paper, we unveil a new vulnerability: the privacy backdoor attack. When a victim fine-tunes a backdoored model, their training data will be leaked at a significantly higher rate than if they had fine-tuned a typical model. Our findings highlight a critical privacy concern within the machine learning community and call for a reevaluation of safety protocols in the use of open-source pre-trained models.
  arXiv  Detail & Related papers  (2024-04-01T16:50:54Z)
• One-bit Flip is All You Need: When Bit-flip Attack Meets Model Training [54.622474306336635]
  A new weight modification attack called bit flip attack (BFA) was proposed, which exploits memory fault inject techniques. We propose a training-assisted bit flip attack, in which the adversary is involved in the training stage to build a high-risk model to release.
  arXiv  Detail & Related papers  (2023-08-12T09:34:43Z)
• Ensemble-based Blackbox Attacks on Dense Prediction [16.267479602370543]
  We show that a carefully designed ensemble can create effective attacks for a number of victim models. In particular, we show that normalization of the weights for individual models plays a critical role in the success of the attacks. Our proposed method can also generate a single perturbation that can fool multiple blackbox detection and segmentation models simultaneously.
  arXiv  Detail & Related papers  (2023-03-25T00:08:03Z)
• Practical No-box Adversarial Attacks against DNNs [31.808770437120536]
  We investigate no-box adversarial examples, where the attacker can neither access the model information or the training set nor query the model. We propose three mechanisms for training with a very small dataset and find that prototypical reconstruction is the most effective. Our approach significantly diminishes the average prediction accuracy of the system to only 15.40%, which is on par with the attack that transfers adversarial examples from a pre-trained Arcface model.
  arXiv  Detail & Related papers  (2020-12-04T11:10:03Z)
• Omni: Automated Ensemble with Unexpected Models against Adversarial Evasion Attack [35.0689225703137]
  A machine learning-based security detection model is susceptible to adversarial evasion attacks. We propose an approach called Omni to explore methods that create an ensemble of "unexpected models" In studies with five types of adversarial evasion attacks, we show Omni is a promising approach as a defense strategy.
  arXiv  Detail & Related papers  (2020-11-23T20:02:40Z)
• Knowledge-Enriched Distributional Model Inversion Attacks [49.43828150561947]
  Model inversion (MI) attacks are aimed at reconstructing training data from model parameters. We present a novel inversion-specific GAN that can better distill knowledge useful for performing attacks on private models from public data. Our experiments show that the combination of these techniques can significantly boost the success rate of the state-of-the-art MI attacks by 150%.
  arXiv  Detail & Related papers  (2020-10-08T16:20:48Z)
• Learning to Attack: Towards Textual Adversarial Attacking in Real-world Situations [81.82518920087175]
  Adversarial attacking aims to fool deep neural networks with adversarial examples. We propose a reinforcement learning based attack model, which can learn from attack history and launch attacks more efficiently.
  arXiv  Detail & Related papers  (2020-09-19T09:12:24Z)
• Improving Robustness to Model Inversion Attacks via Mutual Information Regularization [12.079281416410227]
  This paper studies defense mechanisms against model inversion (MI) attacks. MI is a type of privacy attacks aimed at inferring information about the training data distribution given the access to a target machine learning model. We propose the Mutual Information Regularization based Defense (MID) against MI attacks.
  arXiv  Detail & Related papers  (2020-09-11T06:02:44Z)
• Adversarial Imitation Attack [63.76805962712481]
  A practical adversarial attack should require as little as possible knowledge of attacked models. Current substitute attacks need pre-trained models to generate adversarial examples. In this study, we propose a novel adversarial imitation attack.
  arXiv  Detail & Related papers  (2020-03-28T10:02:49Z)
• DaST: Data-free Substitute Training for Adversarial Attacks [55.76371274622313]
  We propose a data-free substitute training method (DaST) to obtain substitute models for adversarial black-box attacks. To achieve this, DaST utilizes specially designed generative adversarial networks (GANs) to train the substitute models. Experiments demonstrate the substitute models can achieve competitive performance compared with the baseline models.
  arXiv  Detail & Related papers  (2020-03-28T04:28:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.