arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2304.03218cs.LGcs.CV
收藏

Data AUDIT:识别任务模型中由属性效用和可检测性引发的偏差

Data AUDIT: Identifying Attribute Utility- and Detectability-Induced Bias in Task Models

The Johns Hopkins University(约翰斯·霍普金斯大学);Center for Devices and Radiological Health, U.S. Food and Drug Administration(美国食品药品监督管理局器械与放射健康中心)

Mitchell Pavlak, Nathan Drenkow, Nicholas Petrick, Mohammad Mehdi Farhangi, Mathias Unberath

AI总结:

提出了一种针对医学图像数据集的严格定量筛查技术,通过分解数据集属性的可检测性与效用,可靠识别引发模型偏差的隐蔽属性,助力更系统化的算法审计与安全模型部署。

摘要:

为了安全地部署基于深度学习的计算机视觉模型以用于计算机辅助检测和诊断,我们必须确保其稳健性与可靠性。为实现该目标,算法审计受到了广泛关注。为指导审计流程,现有方法依赖于启发式方法或高层目标(例如,在性别、社会性别或种族等受保护属性方面不歧视)。然而,算法可能对超出这些明显属性的各种属性表现出偏差,且与这些更隐蔽属性相关的完整性问题可能产生严重后果。为了生成可操作的、数据驱动的假设,以识别可能引发模型偏差的特定数据集属性,我们提出了一种针对医学图像数据集进行严格定量筛查的首项技术。借鉴因果推断与信息论领域的文献,我们的流程根据数据集属性的可检测性和效用(定义为知晓该属性为任务标签提供的信息量)来分解与其相关的风险。为证明我们方法的有效性与敏感性,我们开发了多种数据集,其中合成了与目标标签具有不同程度关联的伪影,从而允许通过将性能与真实反事实样本进行比较来评估继承的模型偏差。使用这些数据集以及数百个已训练模型的结果,我们表明该筛查方法能可靠地识别几乎不可察觉的引发偏差的伪影。最后,我们将该方法应用于一个流行皮肤病变数据集的自然属性,并证明了其成功。我们的方法提供了一种执行更系统化算法审计的手段,并指导未来的数据收集工作,以追求更安全、更可靠的模型。

英文摘要:

To safely deploy deep learning-based computer vision models for computer-aided detection and diagnosis, we must ensure that they are robust and reliable. Towards that goal, algorithmic auditing has received substantial attention. To guide their audit procedures, existing methods rely on heuristic approaches or high-level objectives (e.g., non-discrimination in regards to protected attributes, such as sex, gender, or race). However, algorithms may show bias with respect to various attributes beyond the more obvious ones, and integrity issues related to these more subtle attributes can have serious consequences. To enable the generation of actionable, data-driven hypotheses which identify specific dataset attributes likely to induce model bias, we contribute a first technique for the rigorous, quantitative screening of medical image datasets. Drawing from literature in the causal inference and information theory domains, our procedure decomposes the risks associated with dataset attributes in terms of their detectability and utility (defined as the amount of information knowing the attribute gives about a task label). To demonstrate the effectiveness and sensitivity of our method, we develop a variety of datasets with synthetically inserted artifacts with different degrees of association to the target label that allow evaluation of inherited model biases via comparison of performance against true counterfactual examples. Using these datasets and results from hundreds of trained models, we show our screening method reliably identifies nearly imperceptible bias-inducing artifacts. Lastly, we apply our method to the natural attributes of a popular skin-lesion dataset and demonstrate its success. Our approach provides a means to perform more systematic algorithmic audits and guide future data collection efforts in pursuit of safer and more reliable models.

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。

↑