Cloud-based AI services offer numerous benefits but also introduce vulnerabilities, allowing for tampering with deployed DNN models, ranging from injecting malicious behaviors to reducing computing resources. Fingerpr...
详细信息
Cloud-based AI services offer numerous benefits but also introduce vulnerabilities, allowing for tampering with deployed DNN models, ranging from injecting malicious behaviors to reducing computing resources. Fingerprint samples are generated to query models to detect such tampering. In this paper, we present Intersecting-Boundary-Sensitive Fingerprinting (IBSF), a novel method for black-box integrity verification of DNN models using only top-1 labels. Recognizing that tampering with a model alters its decision boundary, IBSF crafts fingerprint samples from normal samples by maximizing the partial Shannon entropy of a selected subset of categories to position the fingerprint samples near decision boundaries where the categories in the subset intersect. These fingerprint samples are almost indistinguishable from their source samples. We theoretically establish and confirm experimentally that these fingerprint samples' expected sensitivity to tampering increases with the cardinality of the subset. Extensive evaluation demonstrates that IBSF surpasses existing state-of-the-art fingerprinting methods, particularly with larger subset cardinality, establishing its state-of-the-art performance in black-box tampering detection using only top-1 labels. The IBSF code is available at: https://***/CGCL-codes/IBSF. Copyright 2024 by the author(s)
Machine learning (ML) models are widely deployed on edge nodes, such as mobile phones and edge servers, to power a wide range of AI applications over the web. Ensuring the integrity of these edge models is paramount, ...
详细信息
Emerging byte-addressable non-volatile memory(NVM)technologies offer higher density and lower cost than DRAM,at the expense of lower performance and limited write *** have been many studies on hybrid NVM/DRAM memory m...
详细信息
Emerging byte-addressable non-volatile memory(NVM)technologies offer higher density and lower cost than DRAM,at the expense of lower performance and limited write *** have been many studies on hybrid NVM/DRAM memory management in a single physical ***,it is still an open problem on how to manage hybrid memories efficiently in a distributed *** paper proposes Alloy,a memory resource abstraction and data placement strategy for an RDMA-enabled distributed hybrid memory pool(DHMP).Alloy provides simple APIs for applications to utilize DRAM or NVM resource in the DHMP,without being aware of the hardware details of the *** propose a hotness-aware data placement scheme,which combines hot data migration,data replication and write merging together to improve application performance and reduce the cost of *** evaluate Alloy with several micro-benchmark workloads and public benchmark *** results show that Alloy can significantly reduce the DRAM usage in the DHMP by up to 95%,while reducing the total memory access time by up to 57%compared with the state-of-the-art approaches.
Traditional unlearnable strategies have been proposed to prevent unauthorized users from training on the 2D image data. With more 3D point cloud data containing sensitivity information, unauthorized usage of this new ...
Federated Learning (FL) has emerged as a promising approach for privacy-preserving model training across decentralized devices. However, it faces challenges such as statistical heterogeneity and susceptibility to adve...
详细信息
Machine learning models are increasingly used in time series prediction with promising results. The model explanation of time series prediction falls behind the model development and makes less sense to users in under...
详细信息
Although the containers are featured by light-weightness, it is still resource-consuming to pull and startup a large container image, especially in relatively resource-constrained edge cloud. Fortunately, Docker, as t...
详细信息
Automating the synthesis of User Interfaces (UIs) plays a crucial role in enhancing productivity and accelerating the development lifecycle, reducing both development time and manual effort. Recently, the rapid develo...
详细信息
Blockchain platform Ethereum has involved millions of accounts due to its strong potential for providing numerous services based on smart *** massive accounts can be divided into diverse categories,such as miners,toke...
详细信息
Blockchain platform Ethereum has involved millions of accounts due to its strong potential for providing numerous services based on smart *** massive accounts can be divided into diverse categories,such as miners,tokens,and exchanges,which is termed as account diversity in this *** benefit of investigating diversity are multi-fold,including understanding the Ethereum ecosystem deeper and opening the possibility of tracking certain abnormal ***,the exploration of blockchain account diversity remains *** the most relevant studies,which focus on the deanonymization of the accounts on Bitcoin,can hardly be applied on Ethereum since their underlying protocols and user idioms are *** this end,we present the first attempt to demystify the account diversity on *** key observation is that different accounts exhibit diverse behavior patterns,leading us to propose the heuristics for classification as the *** then raise the coverage rate of classification by the statistical learning model Maximum Likelihood Estimation(MLE).We collect real-world data through extensive efforts to evaluate our proposed method and show its ***,we make an in-depth analysis of the dynamic evolution of the Ethereum ecosystem and uncover the abnormal arbitrage *** for the former,we validate two sweeping statements reliably:(1)standalone miners are gradually replaced by the mining pools and cooperative miners;(2)transactions related to the mining pool and exchanges take up a large share of the total *** latter analysis shows that there are a large number of arbitrage transactions transferring the coins from one exchange to another to make a price difference.
Matrix multiplication (MM) is pivotal in fields from deep learning to scientific computing, driving the quest for improved computational efficiency. Accelerating MM encompasses strategies like complexity reduction, pa...
暂无评论