arXiv subjects

Sébastien Gambs

Publications and source records attributed to Sébastien Gambs.

At least 19 recordsLinked to original sources

arXiv2026

Neighborhood Watch: Privacy Risks in Seeded Local Combination Synthetic Data

Synthetic data is seen as a promising solution for sharing data in sensitive contexts. However, recent work on privacy attacks have shown that there are still significant residual risks, especially for synthetic data generations methods that are not based on formal approaches such as differential privacy. In this paper, we investigate the privacy risks associated with local combination approaches for generating synthetic data in which synthetic profiles are built by combining real neighbouring profiles. More precisely, we focus on three methods from this family, namely SMOTE, Simulant and Avatar, which have been recently used as a way to share 'anonymised data' in the healthcare domain. In particular, we conduct an extensive privacy analysis through a diverse set of attacks: membership inference, linkage and reconstruction attacks. Our results demonstrate substantial privacy leakage for all three methods, raising serious doubts about whether their outputs should be regarded as anonymous in practice.

cs.CR↗

arXiv2026

Manipulation-Proof Oblivious Audits against Deceptive Model Providers

Audits have emerged as a critical instrument for algorithmic governance, providing a mechanism for external scrutiny and governance of machine learning models. However, ensuring the integrity of such assessments remains a challenging issue. For instance in regulatory contexts, audits are typically declared or easily detected, thus enabling model providers to manipulate the process, whether intentionally or inadvertently. This vulnerability is particularly acute in the context of fairness evaluations, in which providers can often infer sensitive attributes and strategically equalize allocation rates between groups to satisfy fairness metrics. In this paper, we introduce a novel audit protocol designed to significantly increase the post-audit detectability of such manipulations by enabling the auditor to query the model in an oblivious manner. Our approach leverages a Private Information Retrieval mechanism to require the provider to label a large set of instances, while preventing it from knowing which subset will ultimately be used for the audit. The protocol is efficient, imposes minimal overhead on the auditor, and requires no modification to the audited model, its training procedure, or its inference pipeline. We provide theoretical guarantees showing that, under this protocol, a provider attempting to hide unfairness must falsify a significantly larger number of responses, thereby increasing both the difficulty and the likelihood of detection of manipulation. Experimental results across representative audit scenarios confirm the effectiveness and practicality of our approach.

cs.LG↗

arXiv2026

Embedding Inference Attack

Embedding models are essential components of modern Information Retrieval (IR) systems, yet they are typically hidden behind APIs. Recent works have shown that dense IR system can lead to security vulnerabilities such as embedding inversion attacks. However, such attacks usually require that the attacker knows the embedding model for the attack to be applicable. In this paper, we study IR systems under a black-box setting in which the adversary observes only the unordered set of retrieved documents, without ranking or similarity scores. We demonstrate that in such contexts, tailored queries allow an adversary to identify which embedding model is in use from a set of known model candidate, which we coin as an embedding inference attack (EIA). We also show that certain queries remain discriminative even when the system includes a reranker as a potential defense mechanism. We further validate our method on a real Retrieval-Augmented Generation (RAG) system, in which the tailored queries bypass the LLM's tendency to reject inputs it does not recognize as well-formed questions. Finally, we propose and evaluate other mitigation strategies such as similarity thresholds.

cs.CR↗

arXiv2026

Revisiting Locally Differentially Private Protocols: Towards Better Trade-offs in Privacy, Utility, and Attack Resistance

Local Differential Privacy (LDP) offers strong privacy protection, especially in settings in which the server collecting the data is untrusted. However, designing LDP mechanisms that achieve an optimal trade-off between privacy, utility and robustness to adversarial inference and integrity attacks remains challenging. In this work, we introduce a general multi-objective optimization framework for refining LDP protocols, enabling the joint optimization of privacy and utility under various adversarial settings. While our framework is flexible to accommodate multiple privacy and security attacks as well as utility metrics, in this paper, we specifically optimize for Attacker Success Rate (ASR) under \emph{data reconstruction attack} as a concrete measure of privacy leakage and Mean Squared Error (MSE) as a measure of utility. Complementarily, we evaluate integrity-oriented threats through data poisoning attacks, providing an additional adversarial perspective. More precisely, we systematically revisit these trade-offs by analyzing eight state-of-the-art LDP frequency estimation protocols and proposing refined counterparts that leverage tailored optimization techniques. Experimental results demonstrate that our proposed adaptive mechanisms consistently outperform their non-adaptive counterparts, achieving substantial reductions in ASR while preserving utility, and pushing closer to the ASR-MSE Pareto frontier. By bridging the gap between theoretical guarantees and real-world vulnerabilities, our framework enables modular and context-aware deployment of LDP mechanisms with tunable privacy-utility-attackability trade-offs.

cs.CR↗

arXiv2026

Robust Federated Learning via Byzantine Filtering over Encrypted Updates

Federated Learning (FL) aims to train a collaborative model while preserving data privacy. However, the distributed nature of this approach still raises privacy and security issues, such as the exposure of sensitive data due to inference attacks and the influence of Byzantine behaviors on the trained model. In particular, achieving both secure aggregation and Byzantine resilience remains challenging, as existing solutions often address these aspects independently. In this work, we propose to address these challenges through a novel approach that combines homomorphic encryption for privacy-preserving aggregation with property-inference-inspired meta-classifiers for Byzantine filtering. First, following the property-inference attacks blueprint, we train a set of filtering meta-classifiers on labeled shadow updates, reproducing a diverse ensemble of Byzantine misbehaviors in FL, including backdoor, gradient-inversion, label-flipping and shuffling attacks. The outputs of these meta-classifiers are then used to cancel the Byzantine encrypted updates by reweighting. Second, we propose an automated method for selecting the optimal kernel and the dimensionality hyperparameters with respect to homomorphic inference, aggregation constraints and efficiency over the CKKS cryptosystem. Finally, we demonstrate through extensive experiments the effectiveness of our approach against Byzantine participants on the FEMNIST, CIFAR10, GTSRB, and acsincome benchmarks. More precisely, our SVM filtering achieves accuracies between $90$% and $94$% for identifying Byzantine updates at the cost of marginal losses in model utility and encrypted inference runtimes ranging from $6$ to $24$ seconds and from $9$ to $26$ seconds for an overall aggregation.

cs.LG↗

arXiv2026

Training Set Reconstruction from Differentially Private Forests: How Effective is DP?

Recent research has shown that structured machine learning models such as tree ensembles are vulnerable to privacy attacks targeting their training data. To mitigate these risks, differential privacy (DP) has become a widely adopted countermeasure, as it offers rigorous privacy protection. In this paper, we introduce a reconstruction attack targeting state-of-the-art $ε$-DP random forests. By leveraging a constraint programming model that incorporates knowledge of the forest's structure and DP mechanism characteristics, our approach formally reconstructs the most likely dataset that could have produced a given forest. Through extensive computational experiments, we examine the interplay between model utility, privacy guarantees and reconstruction accuracy across various configurations. Our results reveal that random forests trained with meaningful DP guarantees can still leak portions of their training data. Specifically, while DP reduces the success of reconstruction attacks, the only forests fully robust to our attack exhibit predictive performance no better than a constant classifier. Building on these insights, we also provide practical recommendations for the construction of DP random forests that are more resilient to reconstruction attacks while maintaining a non-trivial predictive performance.

cs.LG↗

arXiv2025

WaKA: Data Attribution using K-Nearest Neighbors and Membership Privacy Principles

In this paper, we introduce WaKA (Wasserstein K-nearest-neighbors Attribution), a novel attribution method that leverages principles from the LiRA (Likelihood Ratio Attack) framework and k-nearest neighbors classifiers (k-NN). WaKA efficiently measures the contribution of individual data points to the model's loss distribution, analyzing every possible k-NN that can be constructed using the training set, without requiring to sample subsets of the training set. WaKA is versatile and can be used a posteriori as a membership inference attack (MIA) to assess privacy risks or a priori for privacy influence measurement and data valuation. Thus, WaKA can be seen as bridging the gap between data attribution and membership inference attack (MIA) by providing a unified framework to distinguish between a data point's value and its privacy risk. For instance, we have shown that self-attribution values are more strongly correlated with the attack success rate than the contribution of a point to the model generalization. WaKA's different usage were also evaluated across diverse real-world datasets, demonstrating performance very close to LiRA when used as an MIA on k-NN classifiers, but with greater computational efficiency. Additionally, WaKA shows greater robustness than Shapley Values for data minimization tasks (removal or addition) on imbalanced datasets.

cs.LG↗

arXiv2025

On the Usage of Gaussian Process for Efficient Data Valuation

In machine learning, knowing the impact of a given datum on model training is a fundamental task referred to as Data Valuation. Building on previous works from the literature, we have designed a novel canonical decomposition allowing practitioners to analyze any data valuation method as the combination of two parts: a utility function that captures characteristics from a given model and an aggregation procedure that merges such information. We also propose to use Gaussian Processes as a means to easily access the utility function on ``sub-models'', which are models trained on a subset of the training set. The strength of our approach stems from both its theoretical grounding in Bayesian theory, and its practical reach, by enabling fast estimation of valuations thanks to efficient update formulae.

cs.LG↗

arXiv2025

P2NIA: Privacy-Preserving Non-Iterative Auditing

The emergence of AI legislation has increased the need to assess the ethical compliance of high-risk AI systems. Traditional auditing methods rely on platforms' application programming interfaces (APIs), where responses to queries are examined through the lens of fairness requirements. However, such approaches put a significant burden on platforms, as they are forced to maintain APIs while ensuring privacy, facing the possibility of data leaks. This lack of proper collaboration between the two parties, in turn, causes a significant challenge to the auditor, who is subject to estimation bias as they are unaware of the data distribution of the platform. To address these two issues, we present P2NIA, a novel auditing scheme that proposes a mutually beneficial collaboration for both the auditor and the platform. Extensive experiments demonstrate P2NIA's effectiveness in addressing both issues. In summary, our work introduces a privacy-preserving and non-iterative audit scheme that enhances fairness assessments using synthetic or local data, avoiding the challenges associated with traditional API-based audits.

cs.LG↗

arXiv2024

GRAND : Graph Reconstruction from potential partial Adjacency and Neighborhood Data

Cryptographic approaches, such as secure multiparty computation, can be used to compute in a secure manner the function of a distributed graph without centralizing the data of each participant. However, the output of the protocol itself can leak sensitive information about the structure of the original graph. In particular, in this work we propose an approach by which an adversary observing the result of a private protocol for the computation of the number of common neighbors between all pairs of vertices, can reconstruct the adjacency matrix of the graph. In fact, this can only be done up to co-squareness, a notion we introduce, as two different graphs can have the same matrix of common neighbors. We consider two models of adversary, one who observes the common neighbors matrix only, and a knowledgeable one, that has a partial knowledge of the original graph. Our results demonstrate that secure multiparty protocols are not enough for privacy protection, especially in the context of highly structured data such as graphs. The reconstruction that we propose is interesting in itself from the point of view of graph theory.

cs.CR↗

arXiv2024

Smooth Sensitivity for Learning Differentially-Private yet Accurate Rule Lists

Differentially-private (DP) mechanisms can be embedded into the design of a machine learning algorithm to protect the resulting model against privacy leakage. However, this often comes with a significant loss of accuracy due to the noise added to enforce DP. In this paper, we aim at improving this trade-off for a popular class of machine learning algorithms leveraging the Gini impurity as an information gain criterion to greedily build interpretable models such as decision trees or rule lists. To this end, we establish the smooth sensitivity of the Gini impurity, which can be used to obtain thorough DP guarantees while adding noise scaled with tighter magnitude. We illustrate the applicability of this mechanism by integrating it within a greedy algorithm producing rule list models, motivated by the fact that such models remain understudied in the DP literature. Our theoretical analysis and experimental results confirm that the DP rule lists models integrating smooth sensitivity have higher accuracy that those using other DP frameworks based on global sensitivity, for identical privacy budgets.

cs.LG↗

arXiv2024

PANORAMIA: Privacy Auditing of Machine Learning Models without Retraining

We present PANORAMIA, a privacy leakage measurement framework for machine learning models that relies on membership inference attacks using generated data as non-members. By relying on generated non-member data, PANORAMIA eliminates the common dependency of privacy measurement tools on in-distribution non-member data. As a result, PANORAMIA does not modify the model, training data, or training process, and only requires access to a subset of the training data. We evaluate PANORAMIA on ML models for image and tabular data classification, as well as on large-scale language models.

cs.CR↗

arXiv2024

Revealing the True Cost of Locally Differentially Private Protocols: An Auditing Perspective

While the existing literature on Differential Privacy (DP) auditing predominantly focuses on the centralized model (e.g., in auditing the DP-SGD algorithm), we advocate for extending this approach to audit Local DP (LDP). To achieve this, we introduce the LDP-Auditor framework for empirically estimating the privacy loss of locally differentially private mechanisms. This approach leverages recent advances in designing privacy attacks against LDP frequency estimation protocols. More precisely, through the analysis of numerous state-of-the-art LDP protocols, we extensively explore the factors influencing the privacy audit, such as the impact of different encoding and perturbation functions. Additionally, we investigate the influence of the domain size and the theoretical privacy loss parameters $ε$ and $δ$ on local privacy estimation. In-depth case studies are also conducted to explore specific aspects of LDP auditing, including distinguishability attacks on LDP protocols for longitudinal studies and multidimensional data. Finally, we present a notable achievement of our LDP-Auditor framework, which is the discovery of a bug in a state-of-the-art LDP Python package. Overall, our LDP-Auditor framework as well as our study offer valuable insights into the sources of randomness and information loss in LDP protocols. These contributions collectively provide a realistic understanding of the local privacy loss, which can help practitioners in selecting the LDP mechanism and privacy parameters that best align with their specific requirements. We open-sourced LDP-Auditor in \url{https://github.com/hharcolezi/ldp-audit}.

cs.CR↗

arXiv2024

Probabilistic Dataset Reconstruction from Interpretable Models

Interpretability is often pointed out as a key requirement for trustworthy machine learning. However, learning and releasing models that are inherently interpretable leaks information regarding the underlying training data. As such disclosure may directly conflict with privacy, a precise quantification of the privacy impact of such breach is a fundamental problem. For instance, previous work have shown that the structure of a decision tree can be leveraged to build a probabilistic reconstruction of its training dataset, with the uncertainty of the reconstruction being a relevant metric for the information leak. In this paper, we propose of a novel framework generalizing these probabilistic reconstructions in the sense that it can handle other forms of interpretable models and more generic types of knowledge. In addition, we demonstrate that under realistic assumptions regarding the interpretable models' structure, the uncertainty of the reconstruction can be computed efficiently. Finally, we illustrate the applicability of our approach on both decision trees and rule lists, by comparing the theoretical information leak associated to either exact or heuristic learning algorithms. Our results suggest that optimal interpretable models are often more compact and leak less information regarding their training data than greedily-built ones, for a given accuracy level.

cs.AI↗

arXiv2023

SoK: Taming the Triangle -- On the Interplays between Fairness, Interpretability and Privacy in Machine Learning

Machine learning techniques are increasingly used for high-stakes decision-making, such as college admissions, loan attribution or recidivism prediction. Thus, it is crucial to ensure that the models learnt can be audited or understood by human users, do not create or reproduce discrimination or bias, and do not leak sensitive information regarding their training data. Indeed, interpretability, fairness and privacy are key requirements for the development of responsible machine learning, and all three have been studied extensively during the last decade. However, they were mainly considered in isolation, while in practice they interplay with each other, either positively or negatively. In this Systematization of Knowledge (SoK) paper, we survey the literature on the interactions between these three desiderata. More precisely, for each pairwise interaction, we summarize the identified synergies and tensions. These findings highlight several fundamental theoretical and empirical conflicts, while also demonstrating that jointly considering these different requirements is challenging when one aims at preserving a high level of utility. To solve this issue, we also discuss possible conciliation mechanisms, showing that a careful design can enable to successfully handle these different concerns in practice.

cs.LG↗

arXiv2023

Crypto'Graph: Leveraging Privacy-Preserving Distributed Link Prediction for Robust Graph Learning

Graphs are a widely used data structure for collecting and analyzing relational data. However, when the graph structure is distributed across several parties, its analysis is particularly challenging. In particular, due to the sensitivity of the data each party might want to keep their partial knowledge of the graph private, while still willing to collaborate with the other parties for tasks of mutual benefit, such as data curation or the removal of poisoned data. To address this challenge, we propose Crypto'Graph, an efficient protocol for privacy-preserving link prediction on distributed graphs. More precisely, it allows parties partially sharing a graph with distributed links to infer the likelihood of formation of new links in the future. Through the use of cryptographic primitives, Crypto'Graph is able to compute the likelihood of these new links on the joint network without revealing the structure of the private individual graph of each party, even though they know the number of nodes they have, since they share the same graph but not the same links. Crypto'Graph improves on previous works by enabling the computation of a certain number of similarity metrics without any additional cost. The use of Crypto'Graph is illustrated for defense against graph poisoning attacks, in which it is possible to identify potential adversarial links without compromising the privacy of the graphs of individual parties. The effectiveness of Crypto'Graph in mitigating graph poisoning attacks and achieving high prediction accuracy on a graph neural network node classification task is demonstrated through extensive experimentation on a real-world dataset.

cs.CR↗

arXiv2023

Data privacy for Mobility as a Service

Mobility as a Service (MaaS) is revolutionizing the transportation industry by offering convenient, efficient and integrated transportation solutions. However, the extensive use of user data as well as the integration of multiple service providers raises significant privacy concerns. The objective of this survey paper is to provide a comprehensive analysis of the current state of data privacy in MaaS, in particular by discussing the associated challenges, existing solutions as well as potential future directions to ensure user privacy while maintaining the benefits of MaaS systems for society.

cs.CR↗

arXiv2023

On the Risks of Collecting Multidimensional Data Under Local Differential Privacy

The private collection of multiple statistics from a population is a fundamental statistical problem. One possible approach to realize this is to rely on the local model of differential privacy (LDP). Numerous LDP protocols have been developed for the task of frequency estimation of single and multiple attributes. These studies mainly focused on improving the utility of the algorithms to ensure the server performs the estimations accurately. In this paper, we investigate privacy threats (re-identification and attribute inference attacks) against LDP protocols for multidimensional data following two state-of-the-art solutions for frequency estimation of multiple attributes. To broaden the scope of our study, we have also experimentally assessed five widely used LDP protocols, namely, generalized randomized response, optimal local hashing, subset selection, RAPPOR and optimal unary encoding. Finally, we also proposed a countermeasure that improves both utility and robustness against the identified threats. Our contributions can help practitioners aiming to collect users' statistics privately to decide which LDP mechanism best fits their needs.

cs.CR↗