اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدHermes Attack: Steal DNN Models with Lossless Inference Accuracy
174
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Deep Neural Networks (DNNs) models become one of the most valuable enterprise assets due to their critical roles in all aspects of applications. With the trend of privatization deployment of DNN models, the data leakage of the DNN models is becoming increasingly serious and widespread. All existing model-extraction attacks can only leak parts of targeted DNN models with low accuracy or high overhead. In this paper, we first identify a new attack surface -- unencrypted PCIe traffic, to leak DNN models. Based on this new attack surface, we propose a novel model-extraction attack, namely Hermes Attack, which is the first attack to fully steal the whole victim DNN model. The stolen DNN models have the same hyper-parameters, parameters, and semantically identical architecture as the original ones. It is challenging due to the closed-source CUDA runtime, driver, and GPU internals, as well as the undocumented data structures and the loss of some critical semantics in the PCIe traffic. Additionally, there are millions of PCIe packets with numerous noises and chaos orders. Our Hermes Attack addresses these issues by huge reverse engineering efforts and reliable semantic reconstruction, as well as skillful packet selection and order correction. We implement a prototype of the Hermes Attack, and evaluate two sequential DNN models (i.e., MINIST and VGG) and one consequential DNN model (i.e., ResNet) on three NVIDIA GPU platforms, i.e., NVIDIA Geforce GT 730, NVIDIA Geforce GTX 1080 Ti, and NVIDIA Geforce RTX 2080 Ti. The evaluation results indicate that our scheme is able to efficiently and completely reconstruct ALL of them with making inferences on any one image. Evaluated with Cifar10 test dataset that contains 10,000 images, the experiment results show that the stolen models have the same inference accuracy as the original ones (i.e., lossless inference accuracy).
قيم البحث
اقرأ أيضاً
Deep neural networks (DNNs) sustain high performance in todays data processing applications. DNN inference is resource-intensive thus is difficult to fit into a mobile device. An alternative is to offload the DNN inference to a cloud server. However,
such an approach requires heavy raw data transmission between the mobile device and the cloud server, which is not suitable for mission-critical and privacy-sensitive applications such as autopilot. To solve this problem, recent advances unleash DNN services using the edge computing paradigm. The existing approaches split a DNN into two parts and deploy the two partitions to computation nodes at two edge computing tiers. Nonetheless, these methods overlook collaborative device-edge-cloud computation resources. Besides, previous algorithms demand the whole DNN re-partitioning to adapt to computation resource changes and network dynamics. Moreover, for resource-demanding convolutional layers, prior works do not give a parallel processing strategy without loss of accuracy at the edge side. To tackle these issues, we propose D3, a dynamic DNN decomposition system for synergistic inference without precision loss. The proposed system introduces a heuristic algorithm named horizontal partition algorithm to split a DNN into three parts. The algorithm can partially adjust the partitions at run time according to processing time and network conditions. At the edge side, a vertical separation module separates feature maps into tiles that can be independently run on different edge nodes in parallel. Extensive quantitative evaluation of five popular DNNs illustrates that D3 outperforms the state-of-the-art counterparts up to 3.4 times in end-to-end DNN inference time and reduces backbone network communication overhead up to 3.68 times.
It is critical to understand the privacy and robustness vulnerabilities of machine learning models, as their implementation expands in scope. In membership inference attacks, adversaries can determine whether a particular set of data was used in trai
ning, putting the privacy of the data at risk. Existing work has mostly focused on image related tasks; we generalize this type of attack to speaker identification on audio samples. We demonstrate attack precision of 85.9% and recall of 90.8% for LibriSpeech, and 78.3% precision and 90.7% recall for VOiCES (Voices Obscured in Complex Environmental Settings). We find that implementing defenses such as prediction obfuscation, defensive distillation or adversarial training, can reduce attack accuracy to chance.
To ensure protection of the intellectual property rights of DNN models, watermarking techniques have been investigated to insert side-information into the models without seriously degrading the performance of original task. One of the threats for the
DNN watermarking is the pruning attack such that less important neurons in the model are pruned to make it faster and more compact as well as to remove the watermark. In this study, we investigate a channel coding approach to resist the pruning attack. As the channel model is completely different from conventional models like digital images, it has been an open problem what kind of encoding method is suitable for DNN watermarking. A novel encoding approach by using constant weight codes to immunize the effects of pruning attacks is presented. To the best of our knowledge, this is the first study that introduces an encoding technique for DNN watermarking to make it robust against pruning attacks.
An over-the-air membership inference attack (MIA) is presented to leak private information from a wireless signal classifier. Machine learning (ML) provides powerful means to classify wireless signals, e.g., for PHY-layer authentication. As an advers
arial machine learning attack, the MIA infers whether a signal of interest has been used in the training data of a target classifier. This private information incorporates waveform, channel, and device characteristics, and if leaked, can be exploited by an adversary to identify vulnerabilities of the underlying ML model (e.g., to infiltrate the PHY-layer authentication). One challenge for the over-the-air MIA is that the received signals and consequently the RF fingerprints at the adversary and the intended receiver differ due to the discrepancy in channel conditions. Therefore, the adversary first builds a surrogate classifier by observing the spectrum and then launches the black-box MIA on this classifier. The MIA results show that the adversary can reliably infer signals (and potentially the radio and channel information) used to build the target classifier. Therefore, a proactive defense is developed against the MIA by building a shadow MIA model and fooling the adversary. This defense can successfully reduce the MIA accuracy and prevent information leakage from the wireless signal classifier.
Membership inference (MI) attacks affect user privacy by inferring whether given data samples have been used to train a target learning model, e.g., a deep neural network. There are two types of MI attacks in the literature, i.e., these with and with
out shadow models. The success of the former heavily depends on the quality of the shadow model, i.e., the transferability between the shadow and the target; the latter, given only blackbox probing access to the target model, cannot make an effective inference of unknowns, compared with MI attacks using shadow models, due to the insufficient number of qualified samples labeled with ground truth membership information. In this paper, we propose an MI attack, called BlindMI, which probes the target model and extracts membership semantics via a novel approach, called differential comparison. The high-level idea is that BlindMI first generates a dataset with nonmembers via transforming existing samples into new samples, and then differentially moves samples from a target dataset to the generated, non-member set in an iterative manner. If the differential move of a sample increases the set distance, BlindMI considers the sample as non-member and vice versa. BlindMI was evaluated by comparing it with state-of-the-art MI attack algorithms. Our evaluation shows that BlindMI improves F1-score by nearly 20% when compared to state-of-the-art on some datasets, such as Purchase-50 and Birds-200, in the blind setting where the adversary does not know the target models architecture and the target datasets ground truth labels. We also show that BlindMI can defeat state-of-the-art defenses.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
