.png){kind=icon}
返回
Generalized Persistence for Equivariant Operators in Machine Learning
.png){kind=icon}
OA
AI
DOI：10.3390/make5020021.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Artificial neural networks can learn complex, salient data features to achieve a given task. On the opposite end of the spectrum, mathematically grounded methods such as topological data analysis allow users to design analysis pipelines fully aware of data constraints and symmetries. We introduce an original class of neural network layers based on a generalization of topological persistence. The proposed persistence-based layers allow the users to encode specific data properties (e.g., equivariance) easily. Additionally, these layers can be trained through standard optimization procedures (backpropagation) and composed with classical layers. We test the performance of generalized persistence-based layers as pooling operators in convolutional neural networks for image classification on the MNIST, Fashion-MNIST and CIFAR-10 datasets.
Keyword:
topological persistence
rank-based persistence
pooling
steady persistence
期刊
M
IF:
6
论文数:
872
被引数:
1.8K
机构
引用论文
Synthesis, magnetic properties and theoretical calculations of novel nitronyl nitroxide and imino nitroxide diradicals grafted on terpyridine moiety.png)
Polyhedron
PolyhedronIF0
PREAI
分享收藏Persistent-homology-based machine learning: a survey and a comparative study基于持久性同源性的机器学习: 调查和比较研究
PREAI
分享收藏Leaflet Surfaces of Blackspot-resistant and Susceptible Roses and Their Reactions to Fungal InvasionHortScience
OAAI
HortScienceIF0
OAAI分享收藏How health service delivery guides the allocation of major trauma patients in the intensive care units of the inclusive (hub and spoke) trauma system of the Emilia Romagna Region (Italy). A cross-sectional studyBMJ OpenOAAI
BMJ OpenIF0
OAAI分享收藏Oxidative stress in liver of grass carp Ctenopharyngodon idella naturally infected with Saprolegnia parasitica and its influence on disease pathogenesis天然感染水蚤的草鱼肝脏氧化应激及其对疾病发病机制的影响
PREAI
分享收藏
查看更多内容