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 2006.08601cs.LGcs.AI
收藏

解释深度学习中的局部、全局和高阶交互

Explaining Local, Global, And Higher-Order Interactions In Deep Learning

University of Rochester(罗切斯特大学)

Samuel Lerman, Chenliang Xu, Charles Venuto, Henry Kautz

摘要:

我们提出了一个简单而高度概括的方法来解释神经网络的推理过程中的相互作用的部分。首先,我们设计了一个算法的基础上,交叉导数计算统计交互作用的影响,个人的功能,这是广义的2路和高阶(3路或更多)的相互作用。我们提出的结果并排与基于权重的归因技术,证实交叉导数是一个优越的度量2路和高阶交互检测。此外,我们将交叉导数作为神经网络中的解释性设备的使用扩展到计算机视觉设置,通过扩展Grad-CAM,一种流行的基于梯度的CNN解释工具,到高阶。虽然Grad-CAM只能解释图像中单个对象的重要性,但我们称之为Taylor-CAM的方法可以解释神经网络跨多个对象的关系推理。我们展示了我们的解释的成功定性和定量,包括与用户研究。我们将发布所有代码作为工具包,以促进可解释的深度学习。

英文摘要:

We present a simple yet highly generalizable method for explaining interacting parts within a neural network's reasoning process. First, we design an algorithm based on cross derivatives for computing statistical interaction effects between individual features, which is generalized to both 2-way and higher-order (3-way or more) interactions. We present results side by side with a weight-based attribution technique, corroborating that cross derivatives are a superior metric for both 2-way and higher-order interaction detection. Moreover, we extend the use of cross derivatives as an explanatory device in neural networks to the computer vision setting by expanding Grad-CAM, a popular gradient-based explanatory tool for CNNs, to the higher order. While Grad-CAM can only explain the importance of individual objects in images, our method, which we call Taylor-CAM, can explain a neural network's relational reasoning across multiple objects. We show the success of our explanations both qualitatively and quantitatively, including with a user study. We will release all code as a tool package to facilitate explainable deep learning.

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。

↑