Local differential privacy (LDP) provides strict privacy guarantee in a distributed environment. Recent studies demonstrated that LDP protocols are vulnerable to data poisoning attacks where an attacker can manipulate the perturbed result on the local side and send bogus data to skew the final estimate on the server. Unfortunately, existing attack detections do not create an effective attack indicator and rely on particular characteristics of LDP protocols. As a result, they typically exhibit limited detection performance. In this paper, we use log-likelihood as the attack indicator and propose a chain-style detection to enhance the detection effectiveness, in which the attack impact could propagate along the chain and exhibit clear anomaly signal even under stealthy attack scenarios. The experimental results show that our detection consistently outperforms the existing methods. Using four datasets containing categorical and numerical data separately, our detection achieves an F1 score exceeding 96% in most cases. It even remains above 0.9 under stealthy attack settings, outperforming the state-of-the-art detection by up to 0.25.

ABSTRACT Point cloud recognition has wide applications in fields such as autonomous driving and shape classification. Although significant progress has been made in point cloud processing in recent years, most of it has been achieved by designing more complex networks to attain better performance. This paper proposes a novel lightweight point cloud recognition network by introducing a new local neighborhood optimization layer (LNOL), which improves traditional sampling methods by correlation learning in local area. The LNOL is embedded within a single‐layer local transformer architecture, significantly reducing computational complexity and parameters while maintaining the model's expressive power. Experimental results on the ModelNet40 benchmark dataset demonstrate that our method achieves a classification accuracy of 93.3% and an average precision of 92.0% without using a voting strategy. Compared to the mainstream local transformer model point transformer, our network requires only 9.95G FLOPs and 2.33M parameters, reducing computational cost by 94.7% and parameter count by 75.7%, with only a 0.4% drop in accuracy. This study provides an efficient solution for real‐time 3D recognition applications, significantly lowering computational resource requirements while maintaining performance.

Range counting is a fundamental operation in spatial data applications. There is a growing demand to facilitate this operation over a data federation, where spatial data are separately held by multiple data providers (a.k.a., data silos). Most existing data federation schemes employ Secure Multiparty Computation (SMC) to protect privacy, but this approach is computationally expensive and leads to high latency. Consequently, private data federations are often impractical for typical database workloads.This challenge highlights the need for a private data federation scheme capable of providing fast and accurate query responses while maintaining strong privacy. To address this issue, we propose U-DPAP, a utility-aware efficient privacy-preserving method. It is the first scheme to exclusively use differential privacy for privacy protection in spatial data federation, without employing SMC. Moreover, it combines approximate query processing to further enhance efficiency. Our experimental results indicate that a straightforward combination of the two techniques results in unacceptable impacts on data utility. Thus, we design two novel algorithms: one to make differential privacy practical by optimizing the privacy-utility trade-off, and another to address the efficiency-utility trade-off in approximate query processing. The grouping-based perturbation algorithm reduces noise by grouping similar data and applying noise to the groups. The representative data silos selection algorithm minimizes approximate error by selecting representative silos using the similarity between data silos. We rigorously prove the privacy guarantees of U-DPAP. Moreover, experimental results demonstrate that U-DPAP enhances data utility by an order of magnitude while maintaining high communication efficiency.

Personalized federated learning (PFL) is a specialized application of the federated learning paradigm designed to support personalized use cases. Unlike traditional federated learning, which aims to train a high-quality global model, the goal of PFL is to tailor a model that best fits each individual user. Most existing PFL approaches adopt training architectures similar to those used in traditional federated learning, relying on global or partial model sharing during training. While this helps improve model personalization across clients, it also introduces a range of challenges, including risks of data leakage and increased communication overhead. To address these challenges, we propose a novel personalized federated learning (PFL) framework called SuperMPFL, which leverages supermasks to effectively tackle issues related to accuracy, privacy, and efficiency. In particular, the SuperMPFL technique utilizes masking and ranking strategies to obscure the true gradient information. By converting gradients into ranked numerical representations, this approach enhances privacy protection during the training process. Furthermore, this approach reduces communication overhead by transmitting significantly less information compared to conventional methods. In SuperMPFL, each client receives the global model and then emphasizes its personalized parameters, particularly at the model’s edges. This design not only improves accuracy but also strengthens robustness against privacy attacks. Evaluations on standard federated learning benchmarks demonstrate the superiority of our approach, which outperforms state-of-the-art methods in terms of accuracy, privacy, and efficiency.

Adversarial learning is commonly used to extract latent data representations which are expressive to predict the target attribute but indistinguishable in the privacy attribute. However, whether they can achieve an expected privacy-utility tradeoff is of great uncertainty. In this paper, we posit it is the complex interaction between different attributes in the training set that causes disparate tradeoff results. We first formulate the measurement of utility, privacy and their tradeoff in adversarial learning. Then we propose the metrics of Statistical Reliability (SR) and Feature Reliability (FR) to quantify the relationship between attributes. Specifically, SR reflects the co-occurrence sampling bias of the joint distribution between two attributes. Beyond the explicit dependence, FR exploits the intrinsic interaction one attribute exerts on the other via exploring the representation disentanglement. We validate the metrics on CelebA and LFW dataset with a suite of target-privacy attribute pairs. Experimental results demonstrate the strong correlations between the metrics and utility, privacy and their tradeoff. We further conclude how to use SR and FR as a guide to the setting of the privacy-utility tradeoff parameter.

Mechanisms built upon geo-indistinguishability render location privacy, where a user can submit obfuscated locations to Location-Based Service providers but still be able to correctly utilize services. However, these mechanisms are vulnerable under inference attacks. Particularly, with background knowledge of a user’s obfuscated locations, an attacker can infer actual locations by carrying out long-term observation attacks. Unfortunately, how to defend long-term observation attacks in the field of differential location privacy remains open. In this paper, we first demonstrate the vulnerabilities of existing mechanisms under long-term observation attacks. In light of these vulnerabilities, we devise a novel mechanism, referred to as Eclipse, which bridges the gap between location protection and usability of services. Specifically, we harness geo-indistinguishability and$k$-anonymity to obfuscate locations and hide each location based on an anonymity set. As a result, our mechanism effectively perturbs the distribution of locations and suppresses leakage under long-term observation attacks. Moreover, the set of possible outputs is utilized to minimize the impacts to usability and correctness. We formally define and rigorously prove the security of the proposed mechanism by leveraging differential privacy. Moreover, we implement the proposed mechanism and conduct a series of experiments on real-world datasets to demonstrate its efficacy and efficiency.

Users usually have different privacy demands when they contribute individual data to a dataset that is maintained and queried by others. To tackle this problem, several personalized differential privacy (PDP) mechanisms have been proposed to render statistical information of the entire dataset without revealing individual privacy. However, existing mechanisms produce query results with low accuracy, which leads to poor data utility. This is primarily because (1) some users are over protected; (2) utility is not explicitly included in the design objective. Poor data utility impedes the adoption of PDP in the real-world applications. In this paper, we present an adaptive personalized differential privacy framework, called AdaPDP. Specifically, to maximize data utility in different cases, AdaPDP adaptively selects underlying noise generation algorithms and calculates the corresponding parameters based on the type of query functions, data distributions and privacy settings. In addition, AdaPDP performs multiple rounds of utility-aware sampling to satisfy different privacy requirements for users. Our privacy analysis shows that the proposed framework renders rigorous privacy guarantee. We conduct extensive experiments on synthetic and real-world datasets to demonstrate the much less utility losses of the proposed framework over various query functions.

Suffered from the contradiction between the limited capacity of local devices and large size of DNN models, a practical solution is transferring the heavy computational tasks from the local to the server side such as cloud. However, the untrusted server naturally requires all the user data to train neural networks and infer results, which causes the asset loss of the local and raises serious privacy concerns on user's sensitive information. To solve this problem in scenarios of machine learning as a service, we propose a general framework to balance the user privacy, model accuracy and training efficiency, simultaneously. Specifically, our representative subset selection algorithm takes the training value of data into account, selecting the most representative subset from the training data, in order to mitigate the loss of data assets, lower down the transmission overhead from the local to the server and lessen the training burden on the server at the same time. We also design a noisy representation transformation algorithm applying on the features extracted by neural networks to further perturb the data within the selected representative subset. Extensive experiments demonstrate that our framework can run locally with little sacrifice on the computation resource. It can not only protect private data before uploading, but also promote the training efficiency of servers.

Personalized Differential Privacy (PDP) was proposed to satisfy users' different privacy requirements. However, most of the existing PDP mechanisms may significantly destroy the utility of released statistical results. Differentially private statistical results with poor utility may mislead the data analysts, thus it may even decrease the acceptability of the technique used to protect data privacy. Therefore, in this paper, our goal is to pursue higher data utility while satisfying personalized differential privacy. To achieve this goal, we propose the Utility-aware Personalized Exponential Mechanism (UPEM) to effectively achieve PDP while pursuing better utility. UPEM distinguishes the different possible results with the same personalized score, which is used in Personalized Exponential Mechanism (PEM) [1]. PEM considers the personalized privacy budgets of changing elements to achieve PDP. Based on PEM, our UPEM further considers the quantitative changes of these changing tuples to enhance the utility. We confirm the effectiveness and efficiency of UPEM through extensive experiments.

The continuous publication of aggregate statistics over crowd-sourced data to the public has enabled many data mining applications (e.g., real-time traffic analysis). Existing systems usually rely on a trusted server to aggregate the spatio-temporal crowd-sourced data and then apply differential privacy mechanism to perturb the aggregate statistics before publishing to provide strong privacy guarantee. However, the privacy of users will be exposed once the server is hacked or cannot be trusted. In this paper, we study the problem of real-time crowd-sourced statistical data publishing with strong privacy protection under an untrusted server. We propose a novel distributed agent-based privacy-preserving framework, called DADP, that introduces a new level of multiple agents between the users and the untrusted server. Instead of directly uploading the check-in information to the untrusted server, a user can randomly select one agent and upload the check-in information to it with the anonymous connection technology. Each agent aggregates the received crowd-sourced data and perturbs the aggregated statistics locally with Laplace mechanism. The perturbed statistics from all the agents are further combined together to form the entire perturbed statistics for publication. In particular, we propose a distributed budget allocation mechanism and an agent-based dynamic grouping mechanism to realize global w-event ε-differential privacy in a distributed way. We prove that DADP can provide w-event ε-differential privacy for real-time crowd-sourced statistical data publishing under the untrusted server. Extensive experiments on real-world datasets demonstrate the effectiveness of DADP..

Location privacy continues to attract significant attentions from both industry and academia in recent years. However, Location Based Service (LBS) servers or some other adversaries who can monitor a particular user's historical and current status in a long-term way may likely infer user's location privacy. To solve this problem, we propose a Longterm Observation-aware Dummy Selection (LODS) algorithm to achieve k-anonymity for users in LBSs. Different from existing approaches, the LODS takes the historical anonymity sets into account, since mobile users may query LBSs at certain places such as home or office. LODS selects candidate sets containing dummy locations with less number of occurrences firstly, in order to achieve the preferred distribution. Then, LODS further filters out candidate sets with smaller entropy. Finally, we choose the anonymity set with highest Quality of Service (QoS) as the result. Extensive experiment indicates our algorithm can protect user's location privacy effectively against long-term observation, and satisfy user's QoS requirement at the same time.

Heterogeneity among patients always leads to different progression patterns and may require different types of therapy in clinical diagnosis. Therefore, it is crucial to study patient subgrouping. Normally, patient subgrouping is an unsupervised work due to the lack of labeled data. Analysing patients with complex medical data is challenging because of the data multiformity and time irregularity. To handle these issues, we propose a time-sensitive hybrid learning model to subgroup patients. First, we divide the multiform clinical data into two parts: non-time series data and time series data. Then we utilize basic autoencoder (AE) which is a commonly used unsupervised algorithm to learn patients' representations from non-time series data, and we use a recurrent neural network (RNN) based AE to extract representations from time series data. To capture the time irregularity in time series data, we propose a time-sensitive RNN which utilizes the time intervals to control the decaying degree of history memories. Finally, we present a weighted k-means method to subgroup patients with the pairwise representations. Experiments on real world medical datasets demonstrate that our proposed model can effectively improve the validity of patient subgrouping.