arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2009.07974cs.LGcs.AIcs.CVstat.ML
收藏

基于决策边界复杂度的深度神经网络泛化能力分析

Analysis of Generalizability of Deep Neural Networks Based on the Complexity of Decision Boundary

The George Washington University(乔治华盛顿大学)

Shuyue Guan, Murray Loew

摘要:

对于监督学习模型,泛化能力(泛化能力)的分析是至关重要的,因为泛化能力表示模型在未知数据上的表现。传统的泛化方法,如VC维,不适用于深度神经网络(DNN)模型。因此,需要新的理论来解释DNN的普遍性。在这项研究中,我们假设DNN具有更简单的决策边界具有更好的概括性的简约定律(奥卡姆剃刀)。我们创建了决策边界复杂度(DBC)分数来定义和衡量DNN的决策边界的复杂度。DBC分数的思想是在决策边界上或附近生成数据点(称为对抗性示例)。然后,我们的新方法使用这些数据的特征值的熵来测量边界的复杂性。该方法同样适用于高维数据。我们使用训练数据和训练模型来计算DBC分数。而且,模型的推广性的基础事实是它的测试精度。基于DBC分数的实验验证了我们的假设。DBC提供了一种有效的方法来衡量一个决策边界的复杂性,并给出了DNN的泛化能力的定量测量。

英文摘要:

For supervised learning models, the analysis of generalization ability (generalizability) is vital because the generalizability expresses how well a model will perform on unseen data. Traditional generalization methods, such as the VC dimension, do not apply to deep neural network (DNN) models. Thus, new theories to explain the generalizability of DNNs are required. In this study, we hypothesize that the DNN with a simpler decision boundary has better generalizability by the law of parsimony (Occam's Razor). We create the decision boundary complexity (DBC) score to define and measure the complexity of decision boundary of DNNs. The idea of the DBC score is to generate data points (called adversarial examples) on or near the decision boundary. Our new approach then measures the complexity of the boundary using the entropy of eigenvalues of these data. The method works equally well for high-dimensional data. We use training data and the trained model to compute the DBC score. And, the ground truth for model's generalizability is its test accuracy. Experiments based on the DBC score have verified our hypothesis. The DBC is shown to provide an effective method to measure the complexity of a decision boundary and gives a quantitative measure of the generalizability of DNNs.

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。

↑