论文标题
重新考虑基于梯度显着性方法中梯度的阳性聚集和繁殖
Rethinking Positive Aggregation and Propagation of Gradients in Gradient-based Saliency Methods
论文作者
论文摘要
显着性方法通过显示输入元素对该预测的重要性来解释神经网络的预测。一个流行的显着性方法使用梯度信息。在这项工作中，我们从经验上表明，处理梯度信息的两种方法，即积极的聚集和积极的传播，打破了这些方法。尽管这些方法反映了输入中的视觉显着信息，但它们不再解释模型预测，因为生成的显着性图对预测的输出不敏感，并且对模型参数随机化不敏感。专门针对汇总所选层梯度（例如Gradcam ++和FullGrad）的方法，仅聚集了阳性梯度是有害的。我们通过提出几种聚合方法的变体，并通过梯度信息进行积极处理，以进一步支持这一点。对于将诸如LRP，RECTGRAD和引导反向传播等梯度信息进行反向流动信息的方法，我们显示了独家传播正梯度信息的破坏性效果。
Saliency methods interpret the prediction of a neural network by showing the importance of input elements for that prediction. A popular family of saliency methods utilize gradient information. In this work, we empirically show that two approaches for handling the gradient information, namely positive aggregation, and positive propagation, break these methods. Though these methods reflect visually salient information in the input, they do not explain the model prediction anymore as the generated saliency maps are insensitive to the predicted output and are insensitive to model parameter randomization. Specifically for methods that aggregate the gradients of a chosen layer such as GradCAM++ and FullGrad, exclusively aggregating positive gradients is detrimental. We further support this by proposing several variants of aggregation methods with positive handling of gradient information. For methods that backpropagate gradient information such as LRP, RectGrad, and Guided Backpropagation, we show the destructive effect of exclusively propagating positive gradient information.
