Share This Author
AriaNN: Low-Interaction Privacy-Preserving Deep Learning via Function Secret Sharing
- T. Ryffel, Pierre Tholoniat, D. Pointcheval, F. Bach
- Computer ScienceProc. Priv. Enhancing Technol.
- 8 June 2020
This work proposes AriaNN, a low-interaction privacy-preserving framework for private neural network training and inference on sensitive data, and implements the framework as an extensible system on top of PyTorch that leverages CPU and GPU hardware acceleration for cryptographic and machine learning operations.
Formal Verification of Blockchain Byzantine Fault Tolerance
It is argued that it has now become both relatively simple and crucial to formally verify the correctness of blockchain consensus protocols.
Cross-Chain Payment Protocols with Success Guarantees
This paper presents the first cross-chain payment protocol that ensures termination in a bounded amount of time and works correctly in the presence of clock skew, and introduces a new specification formalism called Asynchronous Networks of Timed Automata (ANTA) to formalise such protocols.
Compositional Verification of Byzantine Consensus
A compositional approach to verify a consensus algorithm, for any number n of processes and any upper bound t < n /3 on the number of byzantine processes, and identifies a fairness property that makes this—otherwise purely asynchronous—byzantine consensus algorithm amenable to model checking.
Feasibility of Cross-Chain Payment with Success Guarantees
It is demonstrated that it is possible to solve the problem of cross-chain payment when assuming synchrony, in the sense that each message is guaranteed to arrive within a known amount of time, but impossible to solve withoutAssuming synchrony.
Privacy Budget Scheduling
- Tao Luo, Mingen Pan, Pierre Tholoniat, Asaf Cidon, Roxana Geambasu, M. L'ecuyer
- Computer ScienceOSDI
- 29 June 2021
PrivateKube is described, an extension to the popular Kubernetes datacenter orchestrator that adds privacy as a new type of resource to be managed alongside other traditional compute resources, such as CPU, GPU, and memory.
Holistic Verification of Blockchain Consensus
- N. Bertrand, V. Gramoli, I. Konnov, Marijana Lazi'c, Pierre Tholoniat, Josef Widder
- Computer Science, MathematicsArXiv
- 9 June 2022
This paper formally verify not only the safety properties of the Red Belly Blockchain consensus but also its liveness in about 70 seconds, by model checking for the first time a blockchain consensus used in industry using model checking.
Brief Announcement: Holistic Verification of Blockchain Consensus
- N. Bertrand, V. Gramoli, Igor Konnov, M. Lazic, Pierre Tholoniat, Josef Widder
- Computer SciencePODC
- 20 July 2022
Current progress in mechanical proofs represent the first steps towards verifying blockchain consensus, and the parameterized model checking of threshold automata (TAs) has recently proved instrumental in verifying fully asynchronous parts of consensus algorithms, like broadcast algorithms.