ImPACT: A Networked Service Architecture for Safe Sharing of Restricted Data

Ilya Baldin,Jeff Chase,Jonathan Crabtree, Thomas Nechyba,Laura Christopherson,Michael Stealey,Charley Kneifel,Victor Orlikowski,Rob Carter,Erik Scott,Akio Sone,Don Sizemore

Future Generation Computer Systems（2021）CCF C

RENCI UNC Chapel Hill | Duke Univ | UNC Chapel Hill Davis Lib | Duke Social Sci Res Inst | Duke Univ OIT

Cited 3|Views50

Abstract

In this paper we describe an architecture developed and prototyped in the course of the NSF-funded project called ImPACT—Infrastructure for Privacy-Assured CompuTations. This architecture addresses the common problems that arise from the need to securely store, control access to and process privacy-restricted data in a multi-institutional, multi-stakeholder setting. Specifically the architecture includes several components—a way to publicly advertise a limited set of data attributes without exposing the sensitive data itself; a set of mechanisms for a data owner to specify and automatically enforce complex data-access policies commonly expressed today as Data Use Agreements (DUAs); a way to securely collect digital attestations from multiple stakeholders to satisfy those policies; and a reproducible template to deploy secure processing enclaves in which groups of researchers can analyze the data in a way that complies with data owner policies using the tools of their choice. The paper describes the architecture and its instantiation in a prototype, providing a performance evaluation of several components.

Translated text

Key words

Privacy-restricted data,Data Use Agreement,Authorization logic

求助PDF

上传PDF

Bibtex

AI Read Science

AI Summary

AI Summary is the key point extracted automatically understanding the full text of the paper, including the background, methods, results, conclusions, icons and other key content, so that you can get the outline of the paper at a glance.

Example

Background

Key content

Introduction

Methods

Results

Related work

Fund

Key content

Pretraining has recently greatly promoted the development of natural language processing (NLP)
We show that M6 outperforms the baselines in multimodal downstream tasks, and the large M6 with 10 parameters can reach a better performance
We propose a method called M6 that is able to process information of multiple modalities and perform both single-modal and cross-modal understanding and generation
The model is scaled to large model with 10 billion parameters with sophisticated deployment, and the 10 -parameter M6-large is the largest pretrained model in Chinese
Experimental results show that our proposed M6 outperforms the baseline in a number of downstream tasks concerning both single modality and multiple modalities We will continue the pretraining of extremely large models by increasing data to explore the limit of its performance

Upload PDF to Generate Summary

Must-Reading Tree

Example

Generate MRT to find the research sequence of this paper