Jérémie Decouchant

Jérémie Decouchant

Auto-generated profile, based on publicly available information

Jérémie Decouchant

Delft University of Technology, Delft, Netherlands

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 February 2019 in IEEE Transactions on Computers
Verified
Journal Article
Open access
JYJiangshan Yu
Jiangshan Yu
DKDavid Kozhaya
David Kozhaya
JDJeremie Decouchant
Jeremie Decouchant
PEPaulo Esteves-Verissimo
Paulo Esteves-Verissimo
Existing proof-of-work cryptocurrencies cannot tolerate attackers controlling more than 50% of the network's computing power at any time, but assume that such a condition happening is "unlikely". However, recent attack sophistication, e.g., where attackers can rent mining capacity to obtain a majority of computing power temporarily, render this assumption unrealistic. This paper proposes RepuCoin, the first system to provide guarantees even when more than 50% of the system's computing power is temporarily dominated by an attacker. RepuCoin physically limits the rate of voting power growth of the entire system. In particular, RepuCoin defines a miner's power by its ‘reputation’, as a function of its work integrated over the time of the entire blockchain, rather than through instantaneous computing power, which can be obtained relatively quickly and/or temporarily. As an example, after a single year of operation, RepuCoin can tolerate attacks compromising 51% of the network's computing resources, even if such power stays maliciously seized for almost a whole year. Moreover, RepuCoin provides better resilience to known attacks, compared to existing proof-of-work systems, while achieving a high throughput of 10000 transactions per second (TPS).

28 March 2022
Verified
Conference Paper
Open access
JDJérémie Decouchant
Jérémie Decouchant
DKDavid Kozhaya
David Kozhaya
VRVincent Rahli
Vincent Rahli
JYJiangshan Yu
Jiangshan Yu
Recently, streamlined Byzantine Fault Tolerant (BFT) consensus protocols, such as HotStuff, have been proposed as a means to circumvent the inefficient view-changes of traditional BFT protocols, such as PBFT. Several works have detailed trusted components, and BFT protocols that leverage them to tolerate a minority of faulty nodes and use a reduced number of communication rounds. Inspired by these works we identify two basic trusted services, respectively called the Checker and Accumulator services, which can be leveraged by streamlined protocols. Based on these services, we design Damysus, a streamlined protocol that improves upon HotStuff's resilience and uses less communication rounds. In addition, we show how the Checker and Accumulator services can be adapted to develop Chained-Damysus, a chained version of Damysus where operations are pipelined for efficiency. We prove the correctness of Damysus and Chained-Damysus, and evaluate their performance showcasing their superiority compared to previous protocols.

4 February 2021 in IEEE Transactions on Parallel and Distributed Systems
Verified
Journal Article
Open access
DKDavid Kozhaya
David Kozhaya
JDJeremie Decouchant
Jeremie Decouchant
VRVincent Rahli
Vincent Rahli
PEPaulo Esteves-Verissimo
Paulo Esteves-Verissimo
The accelerated digitalisation of society along with technological evolution have extended the geographical span of cyber-physical systems. Two main threats have made the reliable and real-time control of these systems challenging: (i) uncertainty in the communication infrastructure induced by scale, and heterogeneity of the environment and devices; and (ii) targeted attacks maliciously worsening the impact of the above-mentioned communication uncertainties, disrupting the correctness of real-time applications. This article addresses those challenges by showing how to build distributed protocols that provide both real-time with practical performance, and scalability in the presence of network faults and attacks, in probabilistic synchronous environments. We provide a suite of real-time Byzantine protocols, which we prove correct, starting from a reliable broadcast protocol, called PISTIS, up to atomic broadcast and consensus. This suite simplifies the construction of powerful distributed and decentralized monitoring and control applications, including state-machine replication. Extensive empirical simulations showcase PISTIS's robustness, latency, and scalability. For example, PISTIS can withstand message loss (and delay) rates up to 50 percent in systems with 49 nodes and provides bounded delivery latencies in the order of a few milliseconds.

7 November 2022
Verified
Conference Paper
Open access
BCBart Cox
Bart Cox
LCLydia Y. Chen
Lydia Y. Chen
JDJérémie Decouchant
Jérémie Decouchant
Federated Learning (FL) is a popular deep learning approach that prevents centralizing large amounts of data, and instead relies on clients that update a global model using their local datasets. Classical FL algorithms use a central federator that, for each training round, waits for all clients to send their model updates before aggregating them. In practical deployments, clients might have different computing powers and network capabilities, which might lead slow clients to become performance bottlenecks. Previous works have suggested to use a deadline for each learning round so that the federator ignores the late updates of slow clients, or so that clients send partially trained models before the deadline. To speed up the training process, we instead propose Aergia, a novel approach where slow clients (i) freeze the part of their model that is the most computationally intensive to train; (ii) train the unfrozen part of their model; and (iii) offload the training of the frozen part of their model to a faster client that trains it using its own dataset. The offloading decisions are orchestrated by the federator based on the training speed that clients report and on the similarities between their datasets, which are privately evaluated thanks to a trusted execution environment. We show through extensive experiments that Aergia maintains high accuracy and significantly reduces the training time under heterogeneous settings by up to 27% and 53% compared to FedAvg and TiFL, respectively.

24 September 2018 in IEEE Transactions on Computers
Verified
Journal Article
Open access
DKDavid Kozhaya
David Kozhaya
JDJeremie Decouchant
Jeremie Decouchant
PEPaulo Esteves-Verissimo
Paulo Esteves-Verissimo
Today's cyber-physical systems face various impediments in achieving their intended goals. Namely, communication uncertainties and faults, relative to the increased integration of networked and wireless devices, hinder the synchronism needed to meet real-time deadlines. Moreover, being critical, these systems have higher security threats that cause physical damage. This paper addresses the aforementioned problems by studying how to build the first real-time Byzantine reliable broadcast (RTBRB) protocol tolerating network uncertainties, faults, and attacks. Previous literature contains either real-time reliable broadcast protocols, or asynchronous (non real-time) Byzantine ones.

1 October 2014
Verified
Conference Paper
Open access
SMSonia Ben Mokhtar
Sonia Ben Mokhtar
JDJeremie Decouchant
Jeremie Decouchant
VQVivien Quema
Vivien Quema
Gossip-based content dissemination protocols are a scalable and cheap alternative to centralized content sharing systems. However, it is well known that these protocols suffer from rational nodes, i.e., nodes that aim at downloading the content without contributing their fair share to the system. While the problem of rational nodes that act individually has been well addressed in the literature, colluding rational nodes is still an open issue. Indeed, LiFTinG, the only existing gossip protocol addressing this issue, yields a high ratio of false positive accusations of correct nodes. In this paper, we propose AcTinG, a protocol that prevents rational collusions in gossip-based content dissemination protocols, while guaranteeing zero false positive accusations. We assess the performance of AcTinG on a testbed comprising 400 nodes running on 100 physical machines, and compare its behaviour in the presence of colluders against two state-of-the-art protocols: BAR Gossip that is the most robust protocol handling non-colluding rational nodes, and LiFTinG, the only existing gossip protocol that handles colluding nodes. The performance evaluation shows that AcTinG is able to deliver all messages despite the presence of colluders, whereas LiFTinG and BAR Gossip, both suffer heavy message losses. Finally, using simulations involving up to a million nodes, we show that AcTinG exhibits similar scalability properties as standard gossip-based dissemination protocols.