arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.14601cs.LGcs.CR
收藏

谁怕托马斯·贝叶斯?

Who's Afraid of Thomas Bayes?

Drexel University(德雷塞尔大学);Rapid7(Rapid7公司)

Erick Galinkin

摘要:

在许多情况下,神经网络在测试数据上表现良好,但倾向于高估其对分布外数据的信心。这导致了贝叶斯神经网络的采用,它可以更好地捕捉不确定性,因此更准确地反映模型的置信度。对于机器学习安全研究人员来说,这就提出了一个自然的问题,即贝叶斯模型如何影响模型的安全性。在这项工作中,我们探讨了贝叶斯主义和两种安全措施之间的相互作用:模型隐私和对抗鲁棒性。我们证明了贝叶斯神经网络更容易受到一般的成员推断攻击,但至少与对抗性示例的非贝叶斯对应物一样鲁棒。

英文摘要:

In many cases, neural networks perform well on test data, but tend to overestimate their confidence on out-of-distribution data. This has led to adoption of Bayesian neural networks, which better capture uncertainty and therefore more accurately reflect the model's confidence. For machine learning security researchers, this raises the natural question of how making a model Bayesian affects the security of the model. In this work, we explore the interplay between Bayesianism and two measures of security: model privacy and adversarial robustness. We demonstrate that Bayesian neural networks are more vulnerable to membership inference attacks in general, but are at least as robust as their non-Bayesian counterparts to adversarial examples.

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。

↑