User Tools

Site Tools


9a.018.UVA - Improving Model Performance via Private Multi-Party Leaning of Hidden Features

Project - Summary

This project is inactive.

  • Many domains would benefit from institutions being able to share sensitive material for improved predictive modeling, but parties may be unwilling or unable to do so for legal, ethical, or competitive considerations, among concerns
  • One potential solution is the generation and sharing of synthetic data, which may also contain variables or features that were previously unrecorded by one party or the other – for example two companies or hospitals may not maintain identical patient or customer records, or equipment logs.
  • We propose to use Generative Adversarial Networks (GANs) including a variant known as conditional GANs or cGANs, to investigate the feasibility of transferring synthetic samples of these unrecorded features between parties, and their resultant impact on predictive model performance

Project - Team

Team Member Role Email Phone Number Academic Site/IAB
Peter Beling PI (804) 982-2066 University of Virginia
Stephen Adams Researcher (757) 870-4954 University of Virginia
Alex Langevin Student (412) 339-4904 University of Virginia
Funded by: CGI

Project - Novelty of Approach

Previous work explicitly assumed that cooperating parties recorded the same customer information

This project seeks to relax that assumption and examine the possibility of using GANs to estimate/recover previously unseen or unrecorded information

To the best of our knowledge this is an unexplored application of GANs

Project - Deliverables

1 Experimental results in the form of a CVDI year-end report or academic journal/conference manuscript
2 Code base for modeling of any non-confidential datasets if requested

Project - Benefits to IAB

  • Improved predictive modeling
  • Better business intelligence on customers, patients, equipment, or other areas for improved decision making

Project - Documents

projects/year9/9a.018.uva.txt · Last modified: 2022/10/19 09:04 by sally.johnson