基于描述的可解释性：通过CLIP探测CNN中的偏见
Caption-Driven Explainability: Probing CNNs for Bias via CLIP
AI总结：
本文提出一种基于描述的XAI方法，通过CLIP模型探测CNN中的偏见，以提高模型鲁棒性。
摘要：
鲁棒性已成为机器学习（ML）中最关键的问题之一。将ML模型解释为科学，以理解其行为并提高其鲁棒性，被称为可解释人工智能（XAI）。对于计算机视觉问题，最先进的XAI方法之一是生成显著性图。显著性图突出显示图像中使ML模型最兴奋的像素空间。然而，如果重叠像素空间中存在虚假且显著的特征，这一属性可能会产生误导。在本文中，我们提出了一种基于描述的XAI方法，该方法通过一种新的网络手术方法，将需要解释的独立模型整合到对比语言-图像预训练（CLIP）模型中。所得到的基于描述的XAI模型识别出对模型预测贡献最大的主导概念。这种解释最小化了独立模型受到协变量偏移的风险，并在开发鲁棒的ML模型方面做出了重要贡献。我们的代码可在https://github.com/patch0816/caption-driven-xai获取。
英文摘要：
Robustness has become one of the most critical problems in machine learning (ML). The science of interpreting ML models to understand their behavior and improve their robustness is referred to as explainable artificial intelligence (XAI). One of the state-of-the-art XAI methods for computer vision problems is to generate saliency maps. A saliency map highlights the pixel space of an image that excites the ML model the most. However, this property could be misleading if spurious and salient features are present in overlapping pixel spaces. In this paper, we propose a caption-based XAI method, which integrates a standalone model to be explained into the contrastive language-image pre-training (CLIP) model using a novel network surgery approach. The resulting caption-based XAI model identifies the dominant concept that contributes the most to the models prediction. This explanation minimizes the risk of the standalone model falling for a covariate shift and contributes significantly towards developing robust ML models. Our code is available at https://github.com/patch0816/caption-driven-xai
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。
