1,720,965 research outputs found
Revisiting Stream-Cipher-Based Homomorphic Transciphering in the TFHE Era
No full textInternational audienc
FairCognizer: A model for accurate predictions with inherent fairness evaluation (extended abstract)
No full textInternational audienceAlgorithmic fairness is a critical challenge in build-1 ing trustworthy Machine Learning (ML) mod-2 els. ML classifiers strive to make predictions 3 that closely match real-world observations (ground 4 truth). However, if the ground truth data itself 5 reflects biases against certain sub-populations, a 6 dilemma arises: prioritize fairness and potentially 7 reduce accuracy, or emphasize accuracy at the ex-8 pense of fairness. This work proposes a novel train-9 ing framework that goes beyond achieving high ac-10 curacy. Our framework trains a classifier to not 11 only deliver optimal predictions but also to identify 12 potential fairness risks associated with each predic-13 tion. To do so, we specify a dual-labeling strategy 14 where the second label contains a per-prediction 15 fairness evaluation, referred to as an unfairness 16 risk evaluation. In addition, we identify a sub-17 set of samples as highly vulnerable to group-unfair 18 classifiers. Our experiments demonstrate that our 19 classifiers attain optimal accuracy levels on both 20 the Adult-Census-Income and Compas-Recidivism 21 datasets. Moreover, they identify unfair predictions 22 with nearly 75% accuracy at the cost of expanding 23 the size of the classifier by 45
Optimized Stream-Cipher-Based Transciphering by Means of Functional-Bootstrapping
No full textInternational audienceFully homomorphic encryption suffers from a large expansion in the size of encrypted data, which makes FHE impractical for low-bandwidth networks. Fortunately, transciphering allows to circumvent this issue by involving a symmetric cryptosystem which does not carry the disadvantage of a large expansion factor, and maintains the ability to recover an FHE ciphertext with the cost of extra homomorphic computations on the receiver side. Recent works have started to investigate the efficiency of TFHE as the FHE layer in transciphering, combined with various symmetric schemes including a NIST finalist for lightweight cryptography, namely Grain128-AEAD. Yet, this has so far been done without taking advantage of TFHE functional bootstrapping abilities, that is, evaluating any discrete function “for free” within the bootstrapping operation. In this work, we thus investigate the use of TFHE functional bootstrapping for implementing Grain128-AEAD in a more efficient base (B>2) representation, rather than a binary one. This significantly reduces the overall number of necessary bootstrappings in a homomorphic run of the stream-cipher, for example reducing the number of bootstrappings required in the warm-up phase by a factor of ~3 when B=16
Unveiling the (in) security of threshold FHE-based federated learning: the practical impact of recent CPA D attacks
No full textInternational audienceThe security of Fully Homomorphic Encryption (FHE) has received a lot of attention in recent years with new security notions emerging to better understand the practical attacks that may threaten the real-world deployments of passively secure FHE schemes. One such new notions is CPA D a slight extension of CPA security modelling a passive adversary who is granted access to a decryption oracle accepting only wellformed ciphertexts. While successful CPA D attacks have initially been performed on approximate FHE schemes such as CKKS, recent works have also demonstrated practical CPA D attacks on all mainstream non-approximate FHE, such as BFV, BGV or TFHE. Despite their clear computational practicality, these latter attacks however focus on the abstract security game defining CPA D security. In this paper, we show how to concretely build on these to mount successful FHE key recovery attacks in the Federated Learning (FL) setting, an application scenario of choice for FHE techniques. In FL, participating entities or workers encrypt successive model updates based on their local training data, enabling a central server to aggregate them in order to homomorphically update a global model. As this paper demonstrates, this environment provides a playground for an attacker to launch key recovery attacks against the FHE underlying the secure aggregation mechanism. As such, our findings reveal substantial stealthy key-recovery threats from both the server and a single worker, with very limited impact on the FL training progression or final model qualit
FairCognizer: a model for accurate predictions with inherent fairness evaluation
No full textInternational audienceAlgorithmic fairness is a critical challenge in building trustworthy Machine Learning (ML) models. ML classifiers strive to make predictions that closely match real-world observations (ground truth). However, if the ground truth data itself reflects biases against certain sub-populations, a dilemma arises: prioritize fairness and potentially reduce accuracy, or emphasize accuracy at the expense of fairness. This work proposes a novel training framework that goes beyond achieving high accuracy. Our framework trains a classifier to not only deliver optimal predictions but also to identify potential fairness risks associated with each prediction. To do so, we specify a dual-labeling strategy where the second label contains a per-prediction fairness evaluation, referred to as an unfairness risk evaluation. In addition, we identify a subset of samples as highly vulnerable to group-unfair classifiers. Our experiments demonstrate that our classifiers attain optimal accuracy levels on both the Adult-Census-Income and Compas-Recidivism datasets. Moreover, they identify unfair predictions with nearly 75% accuracy at the cost of expanding the size of the classifier by a mere 45%
Fair play for individuals, foul play for groups? Auditing anonymization’s impact on ML fairness
No full textInternational audienceMachine learning (ML) algorithms are heavily based on the availability of training data, which, depending on the domain, often includes sensitive information about data providers. This raises critical privacy concerns. Anonymization techniques have emerged as a practical solution to address these issues by generalizing features or suppressing data to make it more difficult to accurately identify individuals. Although recent studies have shown that privacy-enhancing technologies can influence ML predictions across different subgroups, thus affecting fair decision-making, the specific effects of anonymization techniques, such as k-anonymity, ℓ-diversity, and t-closeness, on ML fairness remain largely unexplored. In this work, we systematically audit the impact of anonymization techniques on ML fairness, evaluating both individual and group fairness. Our quantitative study reveals that anonymization can degrade group fairness metrics by up to fourfold. Conversely, similarity-based individual fairness metrics tend to improve under stronger anonymization, largely as a result of increased input homogeneity. By analyzing varying levels of anonymization across diverse privacy settings and data distributions, this study provides critical insights into the trade-offs between privacy, fairness, and utility, offering actionable guidelines for responsible AI development. Our code is publicly available at: https://github.com/hharcolezi/anonymity-impact-fairness
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Get PDF“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Get PDFWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
- …
