BriefGPT.xyz
Apr, 2019
序列到序列模型的成员推断攻击: 我的数据是否在您的机器翻译系统中?
Membership Inference Attacks on Sequence-to-Sequence Models
HTML
PDF
Sorami Hisamoto, Matt Post, Kevin Duh
TL;DR
研究了在“机器学习即服务”提供商中数据隐私问题，特别关注基于序列生成模型的会员推理攻击，提供了基于最先进的机器翻译模型的公开数据集，并报告了模型是否泄露私人信息的初步结果。
Abstract
data privacy
is an important issue for "machine learning as a service" providers. We focus on the problem of
membership inference attacks
: given a data sample and black-box access to a model's API, determine whet
→