Théo Ryffel
YOU?
Author Swipe
View article: Federated Analysis With Differential Privacy in Oncology Research: Longitudinal Observational Study Across Hospital Data Warehouses
Federated Analysis With Differential Privacy in Oncology Research: Longitudinal Observational Study Across Hospital Data Warehouses Open
Background Federated analytics in health care allows researchers to perform statistical queries on remote datasets without access to the raw data. This method arose from the need to perform statistical analysis on larger datasets collected…
View article: Generating English Synthetic Documents with Clinical Keywords: A Privacy-Sensitive Methodology
Generating English Synthetic Documents with Clinical Keywords: A Privacy-Sensitive Methodology Open
International audience
View article: Large Language Models as Instructors: A Study on Multilingual Clinical Entity Extraction
Large Language Models as Instructors: A Study on Multilingual Clinical Entity Extraction Open
International audience
View article: Cryptography for Privacy-Preserving Machine Learning
Cryptography for Privacy-Preserving Machine Learning Open
The ever growing use of machine learning (ML), motivated by the possibilities it brings to a large number of sectors, is increasingly raising questions because of the sensitive nature of the data that must be used and the lack of transpare…
View article: Differential Privacy Guarantees for Stochastic Gradient Langevin Dynamics
Differential Privacy Guarantees for Stochastic Gradient Langevin Dynamics Open
We analyse the privacy leakage of noisy stochastic gradient descent by modeling Rényi divergence dynamics with Langevin diffusions. Inspired by recent work on non-stochastic algorithms, we derive similar desirable properties in the stochas…
View article: Differential Privacy Guarantees for Stochastic Gradient Langevin Dynamics
Differential Privacy Guarantees for Stochastic Gradient Langevin Dynamics Open
We analyse the privacy leakage of noisy stochastic gradient descent by modeling Rényi divergence dynamics with Langevin diffusions. Inspired by recent work on non-stochastic algorithms, we derive similar desirable properties in the stochas…
View article: AriaNN: Low-Interaction Privacy-Preserving Deep Learning via Function Secret Sharing
AriaNN: Low-Interaction Privacy-Preserving Deep Learning via Function Secret Sharing Open
We propose AriaNN, a low-interaction privacy-preserving framework for private neural network training and inference on sensitive data.
View article: AriaNN: Low-Interaction Privacy-Preserving Deep Learning via Function Secret Sharing
AriaNN: Low-Interaction Privacy-Preserving Deep Learning via Function Secret Sharing Open
We propose A ria NN, a low-interaction privacy-preserving framework for private neural network training and inference on sensitive data. Our semi-honest 2-party computation protocol (with a trusted dealer) leverages function secret sharing…
View article: End-to-end privacy preserving deep learning on multi-institutional medical imaging
End-to-end privacy preserving deep learning on multi-institutional medical imaging Open
Using large, multi-national datasets for high-performance medical imaging AI systems requires innovation in privacy-preserving machine learning so models can train on sensitive data without requiring data transfer. Here we present PriMIA (…
View article: Syft 0.5: A Platform for Universally Deployable Structured Transparency
Syft 0.5: A Platform for Universally Deployable Structured Transparency Open
We present Syft 0.5, a general-purpose framework that combines a core group of privacy-enhancing technologies that facilitate a universal set of structured transparency systems. This framework is demonstrated through the design and impleme…
View article: Privacy-preserving medical image analysis
Privacy-preserving medical image analysis Open
The utilisation of artificial intelligence in medicine and healthcare has led to successful clinical applications in several domains. The conflict between data usage and privacy protection requirements in such systems must be resolved for …
View article: ARIANN: Low-Interaction Privacy-Preserving Deep Learning via Function\n Secret Sharing
ARIANN: Low-Interaction Privacy-Preserving Deep Learning via Function\n Secret Sharing Open
We propose AriaNN, a low-interaction privacy-preserving framework for private\nneural network training and inference on sensitive data. Our semi-honest\n2-party computation protocol (with a trusted dealer) leverages function secret\nsharin…
View article: Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims
Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims Open
With the recent wave of progress in artificial intelligence (AI) has come a growing awareness of the large-scale impacts of AI systems, and recognition that existing regulations and norms in industry and academia are insufficient to ensure…
View article: Partially Encrypted Machine Learning using Functional Encryption
Partially Encrypted Machine Learning using Functional Encryption Open
Machine learning on encrypted data has received a lot of attention thanks to recent breakthroughs in homomorphic encryption and secure multi-party computation. It allows outsourcing computation to untrusted servers without sacrificing priv…
View article: A generic framework for privacy preserving deep learning
A generic framework for privacy preserving deep learning Open
We detail a new framework for privacy preserving deep learning and discuss its assets. The framework puts a premium on ownership and secure processing of data and introduces a valuable representation based on chains of commands and tensors…