Tag: privacy-preserving machine learning

Three doctoral researchers at the Institute of Science and Technology Austria have been awarded Google PhD Fellowships for their research ...

A new hybrid cryptographic protocol combines homomorphic encryption and streamlined secure exponentiation to train non-linear support vector machines across distributed ...

A new framework called OceanWeave combines Ocean Protocol data markets, sharded compute-to-data consensus, and encrypted federated learning, reporting 3,200 TPS ...

Researchers propose CROG and CLIG, two aggregation algorithms that weight client updates by Euclidean gradient distances to stabilize federated learning ...

A simulated federated learning framework combines carbon-budgeted client selection, blockchain auditing, and gradient-boosting ensembles for cardiovascular prediction.

A federated learning framework combining MRI and clinical data classifies Alzheimer's disease with reported accuracy up to 94.1 percent while ...

A new federated learning framework for chest X-ray analysis reports modest accuracy gains over existing methods while aligning explanations across ...

A systematic study of 603,000 configurations shows that optimization algorithms can find Newton–Raphson starting values for homomorphic division and square ...

Machine learning now touches some of the most sensitive information people generate, from medical records to financial transactions, and the ...

A synthetic simulation of scan-to-order transactions found that DS-GNN ranked privacy-leakage paths and estimated risk more accurately than four baselines, ...