arXiv subjects

John Jewell

Publications and source records attributed to John Jewell.

3 recordsLinked to original sources

arXiv2026

MIDST Challenge at SaTML 2025: Membership Inference over Diffusion-models-based Synthetic Tabular data

Synthetic data is often perceived as a silver-bullet solution to data anonymization and privacy-preserving data publishing. Drawn from generative models like diffusion models, synthetic data is expected to preserve the statistical properties of the original dataset while remaining resilient to privacy attacks. Recent developments of diffusion models have been effective on a wide range of data types, but their privacy resilience, particularly for tabular formats, remains largely unexplored. MIDST challenge sought a quantitative evaluation of the privacy gain of synthetic tabular data generated by diffusion models, with a specific focus on its resistance to membership inference attacks (MIAs). Given the heterogeneity and complexity of tabular data, multiple target models were explored for MIAs, including diffusion models for single tables of mixed data types and multi-relational tables with interconnected constraints. MIDST inspired the development of novel black-box and white-box MIAs tailored to these target diffusion models as a key outcome, enabling a comprehensive evaluation of their privacy efficacy. The MIDST GitHub repository is available at https://github.com/VectorInstitute/MIDST

cs.LG↗

arXiv2023

A Comprehensive View of Personalized Federated Learning on Heterogeneous Clinical Datasets

Federated learning (FL) is increasingly being recognized as a key approach to overcoming the data silos that so frequently obstruct the training and deployment of machine-learning models in clinical settings. This work contributes to a growing body of FL research specifically focused on clinical applications along three important directions. First, we expand the FLamby benchmark (du Terrail et al., 2022a) to include a comprehensive evaluation of personalized FL methods and demonstrate substantive performance improvements over the original results. Next, we advocate for a comprehensive checkpointing and evaluation framework for FL to reflect practical settings and provide multiple comparison baselines. To this end, an open-source library aimed at making FL experimentation simpler and more reproducible is released. Finally, we propose an important ablation of PerFCL (Zhang et al., 2022). This ablation results in a natural extension of FENDA (Kim et al., 2016) to the FL setting. Experiments conducted on the FLamby benchmark and GEMINI datasets (Verma et al., 2017) show that the proposed approach is robust to heterogeneous clinical data and often outperforms existing global and personalized FL techniques, including PerFCL.

cs.LG↗

arXiv2023

Clustering Functional Data via Variational Inference

Functional data analysis deals with data recorded densely over time (or any other continuum) with one or more observed curves per subject. Conceptually, functional data are continuously defined, but in practice, they are usually observed at discrete points. Among different kinds of functional data analyses, clustering analysis aims to determine underlying groups of curves in the dataset when there is no information on the group membership of each individual curve. In this work, we propose a new model-based approach for clustering and smoothing functional data simultaneously via variational inference. We derive coordinate ascent mean-field variational Bayes algorithms to approximate the posterior distribution of our model parameters by finding the variational distribution with the smallest Kullback-Leibler divergence to the posterior. The performance of our proposed method is evaluated using simulated data and publicly available datasets.

stat.ME↗