C4DT - Gestion

Center

Related publications (91)

Understanding the meaning of value to asset-intensive organizations and their stakeholders has been a long -standing goal of researchers and managers of the asset management community. These issues have become particularly prominent with the publication of the ISO 5500x series of standards on asset management. These standards are intended to guide asset-intensive organizations of different types towards realising value from their asset base. However, value can be defined and shaped in various ways, and a high-level conceptual construct to frame these differences is needed. Because such a construct has not yet been established, organizations have no standard guidance to systematically achieve and demonstrate that value is derived from their assets. The paper also discusses the impact of cognitive and motivational biases as well as the complexity of operational and business context while determining the meaning of value. This article presents an overview of theories on value and proposes an original high-level conceptual construct to be used by decision-makers in infrastructure orga-nizations. This holistic proposal provides a structured understanding of the value concept and intends to assist infrastructure asset managers in establishing or developing their decision making processes at all levels of the organization. It enables decision-makers to factor in the benefits and impacts of competing decision options in organizational objectives and stakeholders' needs and expectations and promotes the grasping of opportunities and an enhanced balancing of costs, risks and performance throughout the life cycle of infrastructures.

ELSEVIER SCI LTD2022

System and method for privacy-preserving distributed training of neural network models on distributed datasets

Jean-Philippe Léonard Bossuat, David Jules Froelicher, Jean-Pierre Hubaux, Apostolos Pyrgelis, Sinem Sav, Juan Ramón Troncoso-Pastoriza

A computer-implemented method and a distributed computer system (100) for privacy- preserving distributed training of a global neural network model on distributed datasets (DS1 to DSn). The system has a plurality of data providers (DP1 to DPn) being communicatively coupled. Each data provider has a respective local training dataset (DS1 to DSn) and a vector of output labels (OL1 to OLn) for training the global model. Further, it has a portion of a cryptographic distributed secret key (SK1 to SKn) and a corresponding collective cryptographic public key (CPK) of a multiparty fully homomorphic encryption scheme, with the weights of the global model being encrypted with the collective public key. Each data provider (DP1) computes and aggregates, for each layer of the global model, encrypted local gradients (LG1) using the respective local training dataset (DS1) and output labels (OL1), with forward pass and backpropagation using stochastic gradient descent. At least one data provider homomorphically combines at least a subset of the current local gradients of at least a subset of the data providers into combined local gradients, and updates the weights of the current global model (GM) based on the combined local gradients.

2022

Ultrafast homomorphic encryption models enable secure outsourcing of genotype imputation

Jean-Philippe Léonard Bossuat, David Jules Froelicher, Jean-Pierre Hubaux, Yiping Ma, Juan Ramón Troncoso-Pastoriza

Genotype imputation is a fundamental step in genomic data analysis, where missing variant genotypes are predicted using the existing genotypes of nearby "tag"variants. Although researchers can outsource genotype imputation, privacy concerns may prohibit genetic data sharing with an untrusted imputation service. Here, we developed secure genotype imputation using efficient homomorphic encryption (HE) techniques. In HE-based methods, the genotype data are secure while it is in transit, at rest, and in analysis. It can only be decrypted by the owner. We compared secure imputation with three state-of-the-art non-secure methods and found that HE-based methods provide genetic data security with comparable accuracy for common variants. HE-based methods have time and memory requirements that are comparable or lower than those for the non-secure methods. Our results provide evidence that HE-based methods can practically perform resource-intensive computations for high-throughput genetic data analysis. The source code is freely available for download at https://github.com/K-miran/secure-imputation.

CELL PRESS2021

Privacy and Integrity Preserving Computations with CRISP

Sylvain Chatel, Jean-Pierre Hubaux, Apostolos Pyrgelis, Juan Ramón Troncoso-Pastoriza

In the digital era, users share their personal data with service providers to obtain some utility, e.g., access to high-quality services. Yet, the induced information flows raise privacy and integrity concerns. Consequently, cautious users may want to protect their privacy by minimizing the amount of information they disclose to curious service providers. Service providers are interested in verifying the integrity of the users' data to improve their services and obtain useful knowledge for their business. In this work, we present a generic solution to the trade-off between privacy, integrity, and utility, by achieving authenticity verification of data that has been encrypted for offloading to service providers. Based on lattice-based homomorphic encryption and commitments, as well as zero-knowledge proofs, our construction enables a service provider to process and reuse third-party signed data in a privacy-friendly manner with integrity guarantees. We evaluate our solution on different use cases such as smart-metering, disease susceptibility, and location-based activity tracking, thus showing its versatility. Our solution achieves broad generality, quantum-resistance, and relaxes some assumptions of state-of-the-art solutions without affecting performance.

USENIX ASSOC2021

Efficient Bootstrapping for Approximate Homomorphic Encryption with Non-sparse Keys

Jean-Philippe Léonard Bossuat, Jean-Pierre Hubaux, Christian Vincent Mouchet, Juan Ramón Troncoso-Pastoriza

We present a bootstrapping procedure for the full-RNS variant of the approximate homomorphic-encryption scheme of Cheon et al., CKKS (Asiacrypt 17, SAC 18). Compared to the previously proposed procedures (Eurocrypt 18 & 19, CT-RSA 20), our bootstrapping procedure is more precise, more efficient (in terms of CPU cost and number of consumed levels), and is more reliable and 128-bit-secure. Unlike the previous approaches, it does not require the use of sparse secret-keys. Therefore, to the best of our knowledge, this is the first procedure that enables a highly efficient and precise bootstrapping with a low probability of failure for parameters that are 128-bit-secure under the most recent attacks on sparse R-LWE secrets. We achieve this efficiency and precision by introducing three novel contributions: (i) We propose a generic algorithm for homomorphic polynomial-evaluation that takes into account the approximate rescaling and is optimal in level consumption. (ii) We optimize the key-switch procedure and propose a new technique for linear transformations (double hoisting). (iii) We propose a systematic approach to parameterize the bootstrapping, including a precise way to assess its failure probability. We implemented our improvements and bootstrapping procedure in the open-source Lattigo library. For example, bootstrapping a plaintext in C-32768 takes 18s, has an output coefficient modulus of 505 bits, a mean precision of 19.1 bits, and a failure probability of 2(-)(15)(.)(58). Hence, we achieve 14.1x improvement in bootstrapped throughput (plaintext-bit per second), with respect to the previous best results, and we have a failure probability 468x smaller and ensure 128-bit security.

SPRINGER INTERNATIONAL PUBLISHING AG2021

System and method for privacy-preserving distributed training of machine learning models on distributed datasets

Jean-Philippe Léonard Bossuat, David Jules Froelicher, Joao André Gomes de Sá E Sousa, Jean-Pierre Hubaux, Apostolos Pyrgelis, Sinem Sav, Juan Ramón Troncoso-Pastoriza

A computer-implemented method and a distributed computer system (100) for privacy- preserving distributed training of a global model on distributed datasets (DS1 to DSn). The system has a plurality of data providers (DP1 to DPn) being communicatively coupled. Each data provider has a respective local model (LM1 to LMn) and a respective local training dataset (DS1 to DSn) for training the local model using an iterative training algorithm (IA). Further it has a portion of a cryptographic distributed secret key (SK1 to SKn) and a corresponding collective cryptographic public key (CPK) of a multiparty fully homomorphic encryption scheme, with the local and global model being encrypted with the collective public key. Each data provider (DP1) trains its local model (LM1) using the respective local training dataset (DS1) by executing gradient descent updates of its local model (LM1), and combining (1340) the updated local model (LM1') with the current global model (GM) into a current local model (LM1c). At least one data provider homomorphically combines at least a subset of the current local models of at least a subset of the data providers into a combined model (CM1), and updates the current global model (GM) based on the combined model. The updated global model is provided to at least a subset of the other data providers.

2021

Search for top squark production in fully hadronic final states in proton-proton collisions at root s=13 TeV

Muhammad Ahmad, Georgios Anagnostou, Konstantin Androsov, Tagir Aushev, Thomas Berger, Michele Bianco, Rakesh Chawla, Chunhui Chen, Yixing Chen, Tian Cheng, Pratyush Das, João Miguel das Neves Duarte, Abhisek Datta, Milos Dordevic, Dipanwita Dutta, Matthias Finger, Francesco Fiori, Bertrand François, Sebastiana Gianì, Ruchi Gupta, Seungkyu Ha, Csaba Hajdu, Peter Hansen, Ali Harb, Miao Hu, Muhammad Ansar Iqbal, Alexis Kalogeropoulos, Viktor Khristenko, Doohyun Kim, Donghyun Kim, Ji Hyun Kim, Sanjeev Kumar, Ajay Kumar, Vineet Kumar, Ekaterina Kuznetsova, Federica Legger, Jing Li, Yiming Li, Zhen Liu, Werner Lustermann, Maren Tabea Meinhard, Ioannis Papadopoulos, Anton Petrov, Vladimir Petrov, Jessica Prisciandaro, Andrea Rizzi, Paolo Ronchese, Valérie Scheurer, Sourav Sen, Arvind Shah, Varun Sharma, Ashish Sharma, Wei Shi, Kun Shi, Muhammad Shoaib, Bandeep Singh, Jan Steggemann, Wei Sun, Andromachi Tsirou, David Vannerom, Joao Varela, Yi Wang, Zheng Wang, Xiao Wang, Hui Wang, Jian Wang, Qian Wang, Siyuan Wang, Muhammad Waqas, Matthias Weber, Matthias Wolf, Meng Xiao, Yong Yang, Kai Yi, Hua Zhang, Yi Zhang, Jian Zhao

A search for production of the supersymmetric partners of the top quark, top squarks, is presented. The search is based on proton-proton collision events containing multiple jets, no leptons, and large transverse momentum imbalance. The data were collected with the CMS detector at the CERN LHC at a center-of-mass energy of 13 TeV, and correspond to an integrated luminosity of 137 fb(-1). The targeted signal production scenarios are direct and gluino-mediated top squark production, including scenarios in which the top squark and neutralino masses are nearly degenerate. The search utilizes novel algorithms based on deep neural networks that identify hadronically decaying top quarks and W bosons, which are expected in many of the targeted signal models. No statistically significant excess of events is observed relative to the expectation from the standard model, and limits on the top squark production cross section are obtained in the context of simplified supersymmetric models for various production and decay modes. Exclusion limits as high as 1310 GeVare established at the 95% confidence level on the mass of the top squark for direct top squark production models, and as high as 2260 GeV on the mass of the gluino for gluino-mediated top squark production models. These results represent a significant improvement over the results of previous searches for supersymmetry by CMS in the same final state.

2021

POSEIDON: Privacy-Preserving Federated Neural Network Learning

Jean-Philippe Léonard Bossuat, David Jules Froelicher, Joao André Gomes de Sá E Sousa, Jean-Pierre Hubaux, Apostolos Pyrgelis, Sinem Sav, Juan Ramón Troncoso-Pastoriza

In this paper, we address the problem of privacy-preserving training and evaluation of neural networks in an N-party, federated learning setting. We propose a novel system, POSEIDON, the first of its kind in the regime of privacy-preserving neural network training. It employs multiparty lattice-based cryptography to preserve the confidentiality of the training data, the model, and the evaluation data, under a passive-adversary model and collusions between up to N - 1 parties. To efficiently execute the secure backpropagation algorithm for training neural networks, we provide a generic packing approach that enables Single Instruction, Multiple Data (SIMD) operations on encrypted data. We also introduce arbitrary linear transformations within the cryptographic bootstrapping operation, optimizing the costly cryptographic computations over the parties, and we define a constrained optimization problem for choosing the cryptographic parameters. Our experimental results show that POSEIDON achieves accuracy similar to centralized or decentralized non-private approaches and that its computation and communication overhead scales linearly with the number of parties. POSEIDON trains a 3-layer neural network on the MNIST dataset with 784 features and 60K samples distributed among 10 parties in less than 2 hours.

INTERNET SOC2021

Electron and photon reconstruction and identification with the CMS experiment at the CERN LHC

Muhammad Ahmad, Benjamin William Allen, Georgios Anagnostou, Konstantin Androsov, Tagir Aushev, Thomas Berger, Michele Bianco, Rakesh Chawla, Yixing Chen, Chunhui Chen, Tian Cheng, Pratyush Das, João Miguel das Neves Duarte, Abhisek Datta, Milos Dordevic, Dipanwita Dutta, Matthias Finger, Francesco Fiori, Bertrand François, Sebastiana Gianì, Ruchi Gupta, Seungkyu Ha, Csaba Hajdu, Peter Hansen, Ali Harb, Miao Hu, Muhammad Ansar Iqbal, Alexis Kalogeropoulos, Viktor Khristenko, Doohyun Kim, Donghyun Kim, Ji Hyun Kim, Sanjeev Kumar, Ajay Kumar, Vineet Kumar, Ekaterina Kuznetsova, Federica Legger, Yiming Li, Jing Li, Zhen Liu, Werner Lustermann, Maren Tabea Meinhard, Ioannis Papadopoulos, Anton Petrov, Vladimir Petrov, Jessica Prisciandaro, Quentin Python, Andrea Rizzi, Paolo Ronchese, Valérie Scheurer, Sourav Sen, Arvind Shah, Varun Sharma, Ashish Sharma, Wei Shi, Kun Shi, Muhammad Shoaib, Bandeep Singh, Jan Steggemann, Wei Sun, Andromachi Tsirou, David Vannerom, Joao Varela, Zheng Wang, Jian Wang, Xiao Wang, Hui Wang, Qian Wang, Yi Wang, Siyuan Wang, Matthias Weber, Matthias Wolf, Meng Xiao, Yong Yang, Kai Yi, Yi Zhang, Hua Zhang, Jian Zhao

The performance is presented of the reconstruction and identification algorithms for electrons and photons with the CMS experiment at the LHC. The reported results are based on proton-proton collision data collected at a center-of-mass energy of 13 TeV and recorded in 2016-2018, corresponding to an integrated luminosity of 136 fb(-1). Results obtained from lead-lead collision data collected at root S-NN = 5.02 TeV are also presented. Innovative techniques are used to reconstruct the electron and photon signals in the detector and to optimize the energy resolution. Events with electrons and photons in the final state are used to measure the energy resolution and energy scale uncertainty in the recorded events. The measured energy resolution for electrons produced in Z boson decays in proton-proton collision data ranges from 2 to 5%, depending on electron pseudorapidity and energy loss through bremsstrahlung in the detector material. The energy scale in the same range of energies is measured with an uncertainty smaller than 0.1 (0.3)% in the barrel (endcap) region in proton-proton collisions and better than 1(3)% in the barrel (endcap) region in heavy ion collisions. The timing resolution for electrons from Z boson decays with the full 2016-2018 proton-proton collision data set is measured to be 200 ps.

2021

Search for singly and pair-produced leptoquarks coupling to third-generation fermions in proton-proton collisions at root s=13 TeV

Muhammad Ahmad, Benjamin William Allen, Georgios Anagnostou, Konstantin Androsov, Tagir Aushev, Thomas Berger, Michele Bianco, Rakesh Chawla, Chunhui Chen, Yixing Chen, Tian Cheng, Pratyush Das, João Miguel das Neves Duarte, Abhisek Datta, Milos Dordevic, Dipanwita Dutta, Matthias Finger, Francesco Fiori, Bertrand François, Sebastiana Gianì, Ruchi Gupta, Seungkyu Ha, Csaba Hajdu, Peter Hansen, Ali Harb, Miao Hu, Muhammad Ansar Iqbal, Alexis Kalogeropoulos, Viktor Khristenko, Doohyun Kim, Ji Hyun Kim, Donghyun Kim, Ajay Kumar, Sanjeev Kumar, Vineet Kumar, Ekaterina Kuznetsova, Federica Legger, Jing Li, Yiming Li, Zhen Liu, Werner Lustermann, Maren Tabea Meinhard, Ioannis Papadopoulos, Anton Petrov, Vladimir Petrov, Jessica Prisciandaro, Quentin Python, Andrea Rizzi, Paolo Ronchese, Valérie Scheurer, Sourav Sen, Arvind Shah, Ashish Sharma, Varun Sharma, Wei Shi, Kun Shi, Muhammad Shoaib, Bandeep Singh, Jan Steggemann, Wei Sun, Andromachi Tsirou, David Vannerom, Joao Varela, Zheng Wang, Xiao Wang, Hui Wang, Qian Wang, Siyuan Wang, Jian Wang, Yi Wang, Muhammad Waqas, Matthias Weber, Matthias Wolf, Fan Xia, Meng Xiao, Yong Yang, Kai Yi, Yi Zhang, Hua Zhang, Jian Zhao

A search for leptoquarks produced singly and in pairs in proton-proton collisions is presented. We consider the leptoquark (LQ) to be a scalar particle of charge -1/3e coupling to a top quark plus a tau lepton (t tau) or a bottom quark plus a neutrino (b nu), or a vector particle of charge +2/3e, coupling to t nu or b tau. These choices are motivated by models that can explain a series of anomalies observed in the measurement of B meson decays. In this analysis the signatures t tau nu b and t tau nu are probed, using data recorded by the CMS experiment at the CERN LHC at root s = 13 TeV and that correspond to an integrated luminosity of 137 fb(-1). These signatures have not been previously explored in a dedicated search. The data are found to be in agreement with the standard model prediction. Lower limits at 95% confidence level are set on the LQ mass in the range 0.98-1.73 TeV, depending on the LQ spin and its coupling lambda to a lepton and a quark, and assuming equal couplings for the two LQ decay modes considered. These are the most stringent constraints to date on the existence of leptoquarks in this scenario. (C) 2021 The Author(s). Published by Elsevier B.V.

2021