arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.04391cs.LGcs.CR
收藏

开放世界机器学习的批判性评价

A Critical Evaluation of Open-World Machine Learning

Princeton University(普林斯顿大学)

Liwei Song, Vikash Sehwag, Arjun Nitin Bhagoji, Prateek Mittal

摘要:

开放世界机器学习(ML)将在分布内数据上训练的封闭世界模型与分布外(OOD)检测器相结合,其目的是检测和拒绝OOD输入。以前关于开放世界ML系统的工作通常无法在各种可能的对抗条件下测试其可靠性。因此,在本文中,我们试图了解最先进的开放世界ML系统对系统组件变化的弹性如何?通过对6个OOD检测器的评估,我们发现分布数据、模型架构和OOD数据的选择对OOD检测性能有很大影响,导致误报率超过$70\%$。我们进一步表明,具有22个无意损坏或对抗性扰动的OOD输入使开放世界ML系统无法使用,误报率高达$100\%$。为了提高开放世界ML的弹性,我们将鲁棒分类器与OOD检测技术相结合,并揭示了OOD检测和鲁棒性之间的新权衡。

英文摘要:

Open-world machine learning (ML) combines closed-world models trained on in-distribution data with out-of-distribution (OOD) detectors, which aim to detect and reject OOD inputs. Previous works on open-world ML systems usually fail to test their reliability under diverse, and possibly adversarial conditions. Therefore, in this paper, we seek to understand how resilient are state-of-the-art open-world ML systems to changes in system components? With our evaluation across 6 OOD detectors, we find that the choice of in-distribution data, model architecture and OOD data have a strong impact on OOD detection performance, inducing false positive rates in excess of $70\%$. We further show that OOD inputs with 22 unintentional corruptions or adversarial perturbations render open-world ML systems unusable with false positive rates of up to $100\%$. To increase the resilience of open-world ML, we combine robust classifiers with OOD detection techniques and uncover a new trade-off between OOD detection and robustness.

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。

↑