用概念解释模型为何选某类，还能对比两分类差异。
Concept Based Explanations and Class Contrasting
- 基于概念构建可解释性方法，解析单类预测逻辑。
- 在ResNet50上，91.1%的类别能通过解释找回原判类别。
- 适合想理解模型决策依据的研究者与开发者。
深度神经网络因规模大、非线性强，解释其决策过程极具挑战。本文提出一种基于概念的解释方法，既能说明模型为何预测某一类别，也能对比任意两类之间的差异，即解释为何更倾向于某一类而非另一类。我们在多个公开的ImageNet1K分类模型上进行测试，涵盖定性与定量分析。以PyTorch模型库中的ResNet50为例，利用对类别‘A’的解释，可自动筛选出4个不包含该类别的图像区域，将这些区域拼合后重新输入模型，模型在91.1%的情况下仍正确预测为类别‘A’（覆盖1000个类别中的911个）。相关代码及.ipynb示例已开源至GitHub：https://github.com/rherdt185/concept-based-explanations-and-class-contrasting。
原文摘要 · Abstract (English)
Explaining deep neural networks is challenging, due to their large size and non-linearity. In this paper, we introduce a concept-based explanation method, in order to explain the prediction for an individual class, as well as contrasting any two classes, i.e. explain why the model predicts one class over the other. We test it on several openly available classification models trained on ImageNet1K. We perform both qualitative and quantitative tests. For example, for a ResNet50 model from pytorch model zoo, we can use the explanation for why the model predicts a class 'A' to automatically select four dataset crops where the model does not predict class 'A'. The model then predicts class 'A' again for the newly combined image in 91.1% of the cases (works for 911 out of the 1000 classes). The code including an .ipynb example is available on github: https://github.com/rherdt185/concept-based-explanations-and-class-contrasting
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。