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 2301.08839cs.LG
收藏

一种评估DNN预测的可信度评分

A Trustworthiness Score to Evaluate DNN Predictions

University of Bristol(布里斯托大学);Thales(泰雷兹集团)

Abanoub Ghobrial, Darryl Hond, Hamid Asgari, Kerstin Eder

AI总结:

针对DNN黑箱预测缺乏透明度的问题,提出可信度评分(TS)与可疑度评分(SS),通过检查预测特征量化可信度,在YOLOv5人员检测中提升精确度约20%和5%。

摘要:

由于深度神经网络(DNN)的黑箱特性,在缺乏人工监控的情况下,DNN在运行期间的持续验证具有挑战性。这使得开发人员和监管机构难以对采用DNN的自主系统的部署建立信心。在运行期间,了解DNN的预测是可信的还是可疑的,对于安全性至关重要。在缺乏人工监控的情况下,基本方法是使用模型的输出置信度分数来评估预测是可信还是可疑。然而,模型的置信度分数是来自黑箱计算的结果,因此缺乏透明度,使得难以自动地将可信度赋予预测。我们引入了可信度评分(TS),一种简单的度量,与模型的置信度分数相比,提供了一种更透明、更有效的方式为DNN预测提供信心。该度量通过检查DNN所做预测中某些特征的存在来量化预测的可信度。我们还利用TS度量的基本思想,为整个输入帧提供可疑度评分(SS),以帮助检测存在假阴性的可疑帧。我们使用YOLOv5进行人员检测的案例研究,以展示我们的方法以及TS和SS的使用。案例研究表明,与仅依赖模型置信度分数相比,使用我们的方法在以下两方面持续提高了预测的精确度:1)批准可信预测(约20%的提升)和2)检测可疑帧(约5%的提升)。

英文摘要:

Due to the black box nature of deep neural networks (DNN), the continuous validation of DNN during operation is challenging with the absence of a human monitor. As a result this makes it difficult for developers and regulators to gain confidence in the deployment of autonomous systems employing DNN. It is critical for safety during operation to know when DNN's predictions are trustworthy or suspicious. With the absence of a human monitor, the basic approach is to use the model's output confidence score to assess if predictions are trustworthy or suspicious. However, the model's confidence score is a result of computations coming from a black box, therefore lacks transparency and makes it challenging to automatedly credit trustworthiness to predictions. We introduce the trustworthiness score (TS), a simple metric that provides a more transparent and effective way of providing confidence in DNN predictions compared to model's confidence score. The metric quantifies the trustworthiness in a prediction by checking for the existence of certain features in the predictions made by the DNN. We also use the underlying idea of the TS metric, to provide a suspiciousness score (SS) in the overall input frame to help in the detection of suspicious frames where false negatives exist. We conduct a case study using YOLOv5 on persons detection to demonstrate our method and usage of TS and SS. The case study shows that using our method consistently improves the precision of predictions compared to relying on model confidence score alone, for both 1) approving of trustworthy predictions (~20% improvement) and 2) detecting suspicious frames (~5% improvement).

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。

↑