arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2004.00053cs.LGcs.CLcs.CRstat.ML
收藏

嵌入模型中的信息泄漏

Information Leakage in Embedding Models

Cornell University(康奈尔大学);Google Brain(谷歌大脑);Facebook(脸书)

Congzheng Song, Ananth Raghunathan

摘要:

嵌入是将原始输入数据映射到低维向量表示的函数,同时保留关于输入的重要语义信息。在大量未标记的数据上进行预训练嵌入,并针对下游任务对其进行微调,现在已经成为许多领域实现最先进学习的事实标准。我们证明,嵌入,除了编码通用语义,往往还提出了一个向量,泄漏敏感信息的输入数据。我们开发了三类攻击来系统地研究嵌入可能泄露的信息。首先,嵌入向量可以被反转以部分地恢复某些输入数据。作为一个例子,我们表明,我们对流行的句子嵌入的攻击恢复50 - 70之间的输入单词(F1分数为0.5- 0.7)。第二,嵌入可以揭示输入中固有的敏感属性,并且独立于手头的底层语义任务。文本的作者身份等属性可以通过在少量标记的嵌入向量上训练推理模型来轻松提取。第三,对于不频繁的训练数据输入,嵌入模型会泄漏适量的隶属度信息。我们广泛评估了我们对文本域中各种最先进的嵌入模型的攻击。我们还提出并评估了防御措施,可以防止泄漏在一定程度上在一个较小的成本效用。

英文摘要:

Embeddings are functions that map raw input data to low-dimensional vector representations, while preserving important semantic information about the inputs. Pre-training embeddings on a large amount of unlabeled data and fine-tuning them for downstream tasks is now a de facto standard in achieving state of the art learning in many domains. We demonstrate that embeddings, in addition to encoding generic semantics, often also present a vector that leaks sensitive information about the input data. We develop three classes of attacks to systematically study information that might be leaked by embeddings. First, embedding vectors can be inverted to partially recover some of the input data. As an example, we show that our attacks on popular sentence embeddings recover between 50\%--70\% of the input words (F1 scores of 0.5--0.7). Second, embeddings may reveal sensitive attributes inherent in inputs and independent of the underlying semantic task at hand. Attributes such as authorship of text can be easily extracted by training an inference model on just a handful of labeled embedding vectors. Third, embedding models leak moderate amount of membership information for infrequent training data inputs. We extensively evaluate our attacks on various state-of-the-art embedding models in the text domain. We also propose and evaluate defenses that can prevent the leakage to some extent at a minor cost in utility.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑