基于泰勒系数分析（TCA）的神经网络分类预测解释
Interpreting the predictions of neural network classification based on a Taylor Coefficient Analysis (TCA)
浏览论文内容
中文总结 AI 辅助
本文提出泰勒系数分析（TCA）方法，通过泰勒展开解释神经网络分类预测，在简单任务和LHC实验数据上展示其揭示输入特征影响的能力，并验证二阶TCA足以捕获关键属性。
中文摘要 AI 辅助
我们引入了一个严格且全面的分类体系和范式，用于描述输入特征空间 $X$ 对用于事件分类的神经网络（NN）预测 $\hat{y}$ 的影响，该范式基于 $\hat{y}$ 在 $X$ 中的泰勒展开。我们将这一完整的自省过程称为泰勒系数分析（TCA）。基于两个简单易懂且易于基准测试的示例任务，我们展示了TCA在揭示给定NN模型中$X$的哪些属性导致了$\hat{y}$的何种值方面的强大能力，从而建立对该方法的直觉。一个更复杂的应用旨在代表CERN LHC实验中典型分类任务的$X$。基于该应用，我们演练了TCA提供的不同层次的自省，并讨论了针对CERN LHC数据分析典型的TCA应用中的若干实际问题。最后，我们进行了一项研究，以支持这样的假设：对于CERN LHC实验典型复杂度任务最相关的$X$属性，通常可由TCA捕获至二阶。
英文摘要
We introduce a rigid and comprehensive taxonomy and paradigm for characterizing the influence of the input feature space $X$ on the predictions $\hat{y}$ of a neural network (NN) used for event classification, based on a Taylor expansion of $\hat{y}$ in $X$. The complete process of introspection we refer to as Taylor Coefficient Analysis (TCA). Based on two simplistic example tasks, which can be easily understood and bencmarked, we illustrate the power of the TCA when it comes to revealing, what properties of $X$ have led to what value of $\hat{y}$, of a given NN model, building up intuition for the method. A more complex application is meant to represent $X$ of a typical classification task at a CERN LHC experiment. Based on this application, we play through the different levels of introspection that the TCA offers and discuss a number of practical aspects for a TCA application typical for the analysis of CERN LHC data. We conclude with a study to support the assumption that those properties of $X$ most relevant for tasks of the complexity typical for a CERN LHC experiment, are usually caught by a TCA up to the second order.
发表机构
- Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院）
机构由 AI 辅助整理，请以论文原文为准。
