Accurate training of the Cox proportional hazards model on vertically-partitioned data while preserving privacy

Privacy-Preserving Contrastive Explanations with Local Foil Trees

We present the first algorithm that combines privacy-preserving technologies and state-of-the-art explainable AI to enable privacy-friendly explanations of black-box AI models. We provide a secure algorithm for contrastive explanations of black-box machine learning models that securely trains and uses local foil trees. Our work shows that the...

Distributed Quantum Machine Learning

Quantum computers can solve specific complex tasks for which no reasonable-time classical algorithm is known. Quantum computers do however also offer inherent security of data, as measurements destroy quantum states. Using shared entangled states, multiple parties can collaborate and securely compute quantum algorithms. In this paper we propose...

Efficient Compiler to Covert Security with Public Verifiability for Honest Majority MPC

We present a novel compiler for transforming arbitrary, passively secure MPC protocols into efficient protocols with covert security and public verifiability in the honest majority setting. Our compiler works for protocols with any number of parties > 2 and treats the passively secure protocol in a black-box manner. In multi-party computation ...

Secure Multi-party Computation and Its Applications

There is an urgent need for secure data sharing solutions, such that different organisations can jointly compute with their data, without revealing sensitive data to each other. Secure multi-party computation is an innovative cryptographic technology that has recently become more mature, such that it can be used to obtain very secure data...

Privacy-Preserving Coupling of Vertically-Partitioned Databases and Subsequent Training with Gradient Descent

We show how multiple data-owning parties can collabora tively train several machine learning algorithms without jeopardizing the privacy of their sensitive data. In particular, we assume that every party knows specific features of an overlapping set of people. Using a secure implementation of an advanced hidden set intersection protocol and a...

Technologische doorbraak eindelijk een privacyvriendelijke manier om data te benutten [Technological breakthrough finally, a privacy-friendly way to harness data]

Data sharing and analysis are essential when it comes to achieving economic growth and solving societal challenges. However, data sharing is yet to really get off the ground due to commercial and/or legal barriers, including the fundamental right to privacy. Innovative technologies such as Federated Learning and Multi-Party Computation offer a...

Privacy-preserving dataset combination and Lasso regression for healthcare predictions

Background: Recent developments in machine learning have shown its potential impact for clinical use such as risk prediction, prognosis, and treatment selection. However, relevant data are often scattered across diferent stakehold ers and their use is regulated, e.g. by GDPR or HIPAA. As a concrete use-case, hospital Erasmus MC and health...

Secure integer division with a private divisor

We consider secure integer division within a secret-sharing based secure multi-party computation framework, where the dividend is secret-shared, but the divisor is privately known to a single party. We mention various applications where this situation arises. We give a solution within the passive security model, and extend this to the active...

Coping with the general data protection regulation; anonymization through multi-party computation technology

Analysing combined data sets can result in signifi cant added value for many organisations, but the GDPR has put strict constraints on processing personal data. Anonymization by using Multi-Party Computation (MPC) however may off er organizations some relief of the perceived burden of GDPR under specifi c conditions. In this paper, we will...

Private Hospital Workflow Optimization via Secure k-Means Clustering

Optimizing the workflow of a complex organization such as a hospital is a difficult task. An accurate option is to use a real-time locating system to track locations of both patients and staff. However, privacy regulations forbid hospital management to assess location data of their staff members. In this exploratory work, we propose a secure...

Enabling analytics on sensitive medical data with secure multi-party computation

Secure comparison protocols in the semi-honest model

Due to high complexity, comparison protocols with secret inputs have been a bottleneck in the design of privacy-preserving cryptographic protocols. Different solutions based on homomorphic encryption, garbled circuits and secret sharing techniques have been proposed over the last few years, each claiming high efficiency. Unfortunately, a fair...

Linear round bit-decomposition of secret-shared values

In the field of signal processing in the encrypted domain, linear operations are usually easy to perform, whereas multiplications, and bitwise operations like comparison, are more costly in terms of computation and communication. These bitwise operations frequently require a decomposition of the secret value into bits. To minimize the...

Encrypted integer division and secure comparison

When processing data in the encrypted domain, homomorphic encryption can be used to enable linear operations on encrypted data. Integer division of encrypted data however requires an additional protocol between the client and the server and will be relatively expensive. We present new solutions for dividing encrypted data in the semi-honest...

Privacy-preserving verification of clinical research

We treat the problem of privacy-preserving statistics verification in clinical research. We show that given aggregated results from statistical calculations, we can verify their correctness efficiently, without revealing any of the private inputs used for the calculation. Our construction is based on the primitive of Secure Multi-Party...

Generating private recommendations in a social trust network

Recommender systems have become increasingly important in e-commerce as they can guide customers with finding personalized services and products. A variant of recommender systems that generates recommendations from a set of trusted people is recently getting more attention in social networks. However, people are concerned about their privacy as...