arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2405.00985cs.LGcs.AImath.OCmath.STstat.TH
收藏

ResNet训练中的渐进式前馈坍缩

Progressive Feedforward Collapse of ResNet Training

Academy of Mathematics and Systems Science, Chinese Academy of Sciences(中国科学院数学与系统科学研究院);School of Mathematical Sciences, University of Chinese Academy of Sciences(中国科学院大学数学科学学院)

Sicong Wang, Kuo Gai, Shihua Zhang

AI总结:

本文提出渐进式前馈坍缩(PFC)猜想,刻画ResNet中间层特征坍缩随深度递增的几何规律,并通过多层无约束特征模型(MUFM)揭示其与输入数据的关系。

摘要:

神经坍缩(NC)是深度神经网络(DNNs)在训练终期阶段表现出的一种简单且对称的现象,此时最后一层特征坍缩到各自的类均值,并形成一个与分类器向量对齐的单纯形等角紧框架。然而,最后一层特征在训练过程中与数据及中间层之间的关系尚未被探索。为此,我们刻画了ResNet中间层的几何结构,并提出一个新的猜想——渐进式前馈坍缩(PFC),该猜想认为坍缩程度在DNNs的前向传播过程中逐渐增加。我们根据带权重衰减的ResNet在终期阶段近似Wasserstein空间中的测地曲线这一性质,推导出一个透明的良训练ResNet模型。在多个数据集上,PFC的度量指标确实随深度单调下降。我们提出一个新的替代模型,即多层无约束特征模型(MUFM),通过最优传输正则项连接中间层。MUFM的最优解与NC不一致,但相对于输入数据更加集中。总体而言,本研究将NC扩展为PFC,以建模中间层的坍缩现象及其对输入数据的依赖性,为分类问题中ResNet的理论理解提供了新的启示。

英文摘要:

Neural collapse (NC) is a simple and symmetric phenomenon for deep neural networks (DNNs) at the terminal phase of training, where the last-layer features collapse to their class means and form a simplex equiangular tight frame aligning with the classifier vectors. However, the relationship of the last-layer features to the data and intermediate layers during training remains unexplored. To this end, we characterize the geometry of intermediate layers of ResNet and propose a novel conjecture, progressive feedforward collapse (PFC), claiming the degree of collapse increases during the forward propagation of DNNs. We derive a transparent model for the well-trained ResNet according to that ResNet with weight decay approximates the geodesic curve in Wasserstein space at the terminal phase. The metrics of PFC indeed monotonically decrease across depth on various datasets. We propose a new surrogate model, multilayer unconstrained feature model (MUFM), connecting intermediate layers by an optimal transport regularizer. The optimal solution of MUFM is inconsistent with NC but is more concentrated relative to the input data. Overall, this study extends NC to PFC to model the collapse phenomenon of intermediate layers and its dependence on the input data, shedding light on the theoretical understanding of ResNet in classification problems.

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。

↑