发现卷积层权重均值具有中心对称性，或为模型泛化关键。
On Symmetries in Convolutional Weights
- 分析各层卷积核均值的对称性分布规律
- 对称性与平移、翻转一致性显著相关
- 适合研究模型归纳偏置与架构设计的读者
我们研究了多种卷积神经网络中各层 k×k 卷积核权重的均值对称性。不同于单个神经元偏好特定方向，内部层的均值核通常关于中心呈现对称。本文探讨该对称性在不同数据集和模型中的出现原因，以及其受特定架构选择的影响。结果表明，对称性与平移不变性、翻转一致性等理想性质密切相关，可能构成卷积神经网络的一种内在归纳偏置。
原文摘要 · Abstract (English)
We explore the symmetry of the mean k x k weight kernel in each layer of various convolutional neural networks. Unlike individual neurons, the mean kernels in internal layers tend to be symmetric about their centers instead of favoring specific directions. We investigate why this symmetry emerges in various datasets and models, and how it is impacted by certain architectural choices. We show how symmetry correlates with desirable properties such as shift and flip consistency, and might constitute an inherent inductive bias in convolutional neural networks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。