A5047171761- Cryptography and Data Security
- Solid State Laser Technologies
- Laser Design and Applications
- Advanced Fiber Laser Technologies
- Laser-Matter Interactions and Applications
- Privacy-Preserving Technologies in Data
- Laser-Plasma Interactions and Diagnostics
- Complexity and Algorithms in Graphs
- Photonic Crystal and Fiber Optics
- Blockchain Technology Applications and Security
- Optical Systems and Laser Technology
- Security and Verification in Computing
- Internet Traffic Analysis and Secure E-voting
- Laser-induced spectroscopy and plasma
- Distributed systems and fault tolerance
- Radioactive contamination and transfer
- Nuclear and radioactivity studies
- Spacecraft and Cryogenic Technologies
- Cryptographic Implementations and Security
- IPv6, Mobility, Handover, Networks, Security
- Geophysics and Sensor Technology
- Photorefractive and Nonlinear Optics
- Graphite, nuclear technology, radiation studies
- Mobile Crowdsensing and Crowdsourcing
- Advanced Numerical Methods in Computational Mathematics
École Polytechnique
2011-2024
Laboratoire pour l'utilisation des lasers intenses
2013-2024
Hong Kong University of Science and Technology
2017-2024
University of Hong Kong
2017-2024
Centre National de la Recherche Scientifique
2010-2024
Sorbonne Université
2013-2024
Commissariat à l'Énergie Atomique et aux Énergies Alternatives
2015-2024
Indian Institute of Technology Hyderabad
2023
University College Dublin
2023
University of Maryland, College Park
2016-2019
In the 2015 review paper ‘Petawatt Class Lasers Worldwide’ a comprehensive overview of current status high-power facilities ${>}200~\text{TW}$ was presented. This largely based on facility specifications, with some description their uses, for instance in fundamental ultra-high-intensity interactions, secondary source generation, and inertial confinement fusion (ICF). With 2018 Nobel Prize Physics being awarded to Professors Donna Strickland Gerard Mourou development technique chirped...
Sensor networks have recently found many popular applications in a number of different settings. Sensors at locations can generate streaming data, which be analyzed real-time to identify events interest. In this paper, we propose framework that computes distributed fashion an approximation multi-dimensional data distributions order enable complex resource-constrained sensor networks.We motivate our technique the context problem outlier detection. We demonstrate how extended either distance-...
The protection of user privacy is an important concern in machine learning, as evidenced by the rolling out General Data Protection Regulation (GDPR) European Union (EU) May 2018. GDPR designed to give users more control over their personal data, which motivates us explore learning frameworks for data sharing that do not violate privacy. To meet this goal, article, we propose a novel lossless privacy-preserving tree-boosting system known SecureBoost setting federated learning. first conducts...
The objective of the Apollon 10 PW project is generation peak power pulses 15 fs at $1~\text{shot}~\text{min}^{-1}$ . In this paper a brief update on current status presented, followed by more detailed presentation our experimental and theoretical investigations temporal characteristics laser. More specifically design considerations as well technological physical limitations to achieve intended pulse duration contrast are discussed.
We study the problem of dynamic symmetric searchable encryption. In that setting, it is crucial to minimize information revealed server as a result update operations (insertions and deletions). Two relevant privacy properties have been defined in context: forward backward privacy. The first makes hard for link an operation with previous queries has extensively studied literature. second limits what can learn about entries were deleted from database, happen after deletion. Backward was...
The protection of user privacy is an important concern in machine learning, as evidenced by the rolling out General Data Protection Regulation (GDPR) European Union (EU) May 2018. GDPR designed to give users more control over their personal data, which motivates us explore learning frameworks for data sharing that do not violate privacy. To meet this goal, paper, we propose a novel lossless privacy-preserving tree-boosting system known SecureBoost setting federated learning. first conducts...
The objective of the Apollon project is generation 10 PW peak power pulses 15 fs at 1 shot/minute. In this paper facility design, technological challenges and current progress will be presented.
Cloud database systems such as Amazon RDS or Google SQLenable the outsourcing of a large to server who then responds SQL queries. A natural problem here is efficiently verify correctness responses returned by (untrusted) server. In this paper we present vSQL, novel cryptographic protocol for publicly verifiable queries on dynamic databases. At high level, our construction relies two extensions CMT interactive-proof [Cormode et al., 2012]: (i) supporting outsourced input via use...
We study the problem of dynamic searchable encryption (DSE) with forward-and-backward privacy.Many DSE schemes have been proposed recently but most efficient ones one limitation: they require maintaining operation counter for each unique keyword, either stored locally at client or encrypted server and accessed obliviously (e.g., an oblivious map) during every operation.We propose three new that overcome above limitation achieve constant permanent storage improved performance, both...
The objective of the "Institut de Lumie`re Extreme" consists in providing early 2014 to international scientific community an unique laser based facility delivering 150 J, 15 fs pulses at 1 shot per minute repetition rate allowing investigate unexplored domain laser-matter interaction 1023 - 1024 W/cm2 intensity level.
We study the problem of verifiable computation (VC) for RAM programs, where a computationally weak verifier outsources execution program to powerful (but untrusted) prover. Existing efficient implementations VC protocols require an expensive preprocessing phase that binds parties single circuit. (While there are schemes avoid entirely, their performance remains significantly worse than constructions with preprocessing.) Thus, prover and forced choose between two approaches: (1) Allow...
We use cryptographic techniques to study zone enumeration in DNSSEC.DNSSEC is designed prevent attackers from tampering with domain name system (DNS) messages.The machinery used DNSSEC, however, also creates a new vulnerability, enumeration, enabling an adversary small number of online DNSSEC queries combined offline dictionary attacks learn which names are present or absent DNS zone.We prove that the current standard, NSEC and NSEC3 records, inherently suffers enumeration: specifically, we...
We study the problem of authenticated multi-dimensional range queries over outsourced databases, where an owner outsources its database to untrusted server, which maintains it and answers clients. Previous schemes either scale exponentially in number query dimensions, or rely on heuristic data structures without provable bounds. Most importantly, existing work requires exponential, attributes, support every possible combination dimensions database. In this paper, we propose first that (i)...
When it comes to anonymizing cryptocurrencies, one size most definitely does not fit all.
Transparency logs allow users to audit a potentially malicious service, paving the way towards more accountable Internet. For example, Certificate (CT) enables domain owners Authorities (CAs) and detect impersonation attacks. Yet, achieve their full potential, transparency must be bandwidth-efficient when queried by users. Specifically, everyone should able efficientlylook up log entries keyand efficiently verify that remainsappend-only. Unfortunately, without additional trust assumptions,...
Data protection regulatory policies, such as the European Union's General Protection Regulation (GDPR), force website operators to request users' consent before collecting any personal information revealed through their web browsing. Website operators, motivated by potential value of collected data, employ various methods when designing notices (e.g., dark patterns) in order convince users allow collection much data possible. In this paper, we design and conduct a user study where 1100 MTurk...
A zero-knowledge proof of training (zkPoT) enables a party to prove that they have correctly trained committed model based on dataset without revealing any additional information about the or dataset. An ideal zkPoT should offer provable security and privacy guarantees, succinct size verifier runtime, practical prover efficiency. In this work, we present Kaizen, targeted for deep neural networks (DNNs) achieves all these goals at once. Our construction iteratively train their via...
We present a high-power diode-pumped Yb:CaF(2) laser operating at cryogenic temperature (77 K). A output power of 97 W 1034 nm was extracted for pump 245 W. The corresponding global extraction efficiency (versus absorbed power) is 65%. small signal gain found to be equal 3.1. wavelength could tuned between 990 and 1052 with peaks that correspond well the structure cross-section spectra registered low temperature.
We demonstrate a simple and robust passive coherent combining technique for temporal compression of millijoule energy laser pulses down to few-cycle duration in gas-filled hollow fiber. High efficiency is achieved by using carefully oriented calcite plates pulse division recombination. Carrier-envelope phase (CEP)-stable, 6-fs, 800-nm with more than 0.6 mJ are routinely generated. This method could aid the scaling CEP-stable hollow-fiber compressor systems.