Tetrad: Actively Secure 4PC for Secure Training and Inference

Koti, Nishat; Patra, Arpita; Rachuri, Rahul; Suresh, Ananda Theertha

doi:10.14722/ndss.2022.24058

Cited by 26 publications

(17 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While there exists a plethora of such PPML techniques utilizing PETs (cf. §A.1), we resort to PPML training and inference using MPC techniques in HyFL (Mohassel & Zhang, 2017;Koti et al, 2022b). In particular, most of these works employ 2-4 MPC servers (Mohassel & Rindal, 2018;Byali et al, 2020;Koti et al, 2021;2022b), while recent works started focusing on more servers (Koti et al, 2022a).…”

Section: A13 Differential Privacy (Dp)mentioning

confidence: 99%

“…§A.1), we resort to PPML training and inference using MPC techniques in HyFL (Mohassel & Zhang, 2017;Koti et al, 2022b). In particular, most of these works employ 2-4 MPC servers (Mohassel & Rindal, 2018;Byali et al, 2020;Koti et al, 2021;2022b), while recent works started focusing on more servers (Koti et al, 2022a). We refer to Cabrero-Holgueras & Pastrana for an overview of various PPML approaches.…”

Section: A13 Differential Privacy (Dp)mentioning

confidence: 99%

See 1 more Smart Citation

HyFL: A Hybrid Approach For Private Federated Learning

Marx¹,

Schneider²,

Suresh³

et al. 2023

Preprint

View full text Add to dashboard Cite

As a distributed machine learning paradigm, federated learning (FL) conveys a sense of privacy to contributing participants because training data never leaves their devices. However, gradient updates and the aggregated model still reveal sensitive information. In this work, we propose HyFL, a new framework that combines private training and inference with secure aggregation and hierarchical FL to provide end-to-end protection and facilitate large-scale global deployments. Additionally, we show that HyFL strictly limits the attack surface for malicious participants: they are restricted to data-poisoning attacks and cannot significantly reduce accuracy.

show abstract

Section: A13 Differential Privacy (Dp)mentioning

confidence: 99%

Section: A13 Differential Privacy (Dp)mentioning

confidence: 99%

HyFL: A Hybrid Approach For Private Federated Learning

Marx¹,

Schneider²,

Suresh³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…In this case, the PPML protocols in the four-party setting have a better performance than those in the three-party setting, but require a stronger assumption about the number of honest parties. Among these PPML protocols achieving fairness or GOD, Tetrad [294] has the best efficiency for now. Particularly, Tetrad takes 183 s and 35 GB of communication to train a VGG-16 model over a small dataset CIFAR-10 that includes 50,000 training samples and 10 different classes.…”

Section: Mpc Application To Machine Learningmentioning

confidence: 99%

“…It is an interesting open problem to construct a concretely efficient edaBits protocol with malicious security, which achieves an overhead of 2 or even smaller. In the honest-majority setting, the protocols against malicious adversaries, which allow to convert between the arithmetic, Boolean, and garbling worlds, can be constructed more efficiently [19,294], where the techniques underlying the constant-round MPC protocols (e.g., [236,239]) can be used and adapted.…”

Section: Mpc Application To Machine Learningmentioning

confidence: 99%

Concretely efficient secure multi-party computation protocols: survey and more

Feng

Yang

2022

Security and Safety

View full text Add to dashboard Cite

Secure multi-party computation (MPC) allows a set of parties to jointly compute a function on their private inputs, and reveals nothing but the output of the function. In the last decade, MPC has rapidly moved from a purely theoretical study to an object of practical interest, with a growing interest in practical applications such as privacy-preserving machine learning (PPML). In this paper, we comprehensively survey existing work on concretely efficient MPC protocols with both semi-honest and malicious security, in both dishonest-majority and honest-majority settings. We focus on considering the notion of security with abort, meaning that corrupted parties could prevent honest parties from receiving output after they receive output. We present high-level ideas of the basic and key approaches for designing different styles of MPC protocols and the crucial building blocks of MPC. For MPC applications, we compare the known PPML protocols built on MPC, and describe the efficiency of private inference and training for the state-of-the-art PPML protocols. Furthermore, we summarize several challenges and open problems to break though the efficiency of MPC protocols as well as some interesting future work that is worth being addressed. This survey aims to provide the recent development and key approaches of MPC to researchers, who are interested in knowing, improving, and applying concretely efficient MPC protocols.

show abstract

“…These advancements significantly enhance the efficiency of MPC making it more and more practical for real-world applications. Due to the practical efficiency it can provide, various works [9,21,26,74,75,102] have recently concentrated on MPC for a small number of parties, especially in the three and four party honest majority setting tolerating one corruption. In RIPPLE, we employ MPC techniques across three servers to enable an anonymous communication channel (cf.…”

Section: Garbled Cuckoo Table (Gct)mentioning

confidence: 99%

Privacy-Preserving Epidemiological Modeling on Mobile Graphs

Günther¹,

Holz²,

Judkewitz³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Over the last two years, governments all over the world have used a variety of containment measures to control the spread of COVID-19, such as contact tracing, social distance regulations, and curfews. Epidemiological simulations are commonly used to assess the impact of those policies before they are implemented in actuality. Unfortunately, their predictive accuracy is hampered by the scarcity of relevant empirical data, concretely detailed social contact graphs. As this data is inherently privacy-critical, there is an urgent need for a method to perform powerful epidemiological simulations on real-world contact graphs without disclosing sensitive information.In this work, we present RIPPLE, a privacy-preserving epidemiological modeling framework that enables the execution of a wide range of standard epidemiological models for any infectious disease on a population's most recent real contact graph while keeping all contact information private locally on the participants' devices. In this regard, we also present PIR-SUM, a novel extension to private information retrieval that allows users to securely download the sum of a set of elements from a database rather than individual elements. Our theoretical constructs are supported by a proof-ofconcept implementation in which we show that a 2-week simulation over a population of half a million can be finished in 7 minutes with each participant consuming less than 50 KB of data.

show abstract

Tetrad: Actively Secure 4PC for Secure Training and Inference

Cited by 26 publications

References 45 publications

HyFL: A Hybrid Approach For Private Federated Learning

HyFL: A Hybrid Approach For Private Federated Learning

Concretely efficient secure multi-party computation protocols: survey and more

Privacy-Preserving Epidemiological Modeling on Mobile Graphs

Contact Info

Product

Resources

About