.png){kind=icon}
返回
Blockwise Recursive Moore-Penrose Inverse for Network Learning
.png){kind=icon}
PRE
AI
DOI：10.1109/TSMC.2021.3064241.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Training neural networks with the Moore-Penrose (MP) inverse has recently gained attention in view of its noniterative training nature. However, a significant drawback of learning based on the MP inverse is that the computational memory consumption grows along with the size of a dataset. In this article, based on the partitioning of the MP inverse, we propose a blockwise recursive MP inverse formulation (BRMP) for network learning with low-memory property while preserving its training effectiveness. The BRMP is an equivalent formulation to its batchwise counterpart since neither approximation nor assumption is made in the derivation process. Our further exploration of this recursive method leads to a switching structure among three different scenarios. This structure also reveals that the well-known recursive least squares method is a special case of our proposed technique. Subsequently, we apply BRMP to the training of radial basis function networks as well as multilayer perceptrons. The experimental validation covers both regression and classification tasks.
Keyword:
Training
Artificial neural networks
Task analysis
Switches
Radial basis function networks
Nonhomogeneous media
Memory management
Blockwise recursive Moore-Penrose (MP) inverse
memory reduction
switching structure
training neural networks (NNs)
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
10.5
论文数:
1.1W
被引数:
5.0W
机构
引用论文
On the performance of air-based solar heating systems utilizing phase-change energy storage.png)
Energy
EnergyIF0
PREAI
分享收藏
Wi-HSNN: A subnetwork-based encoding structure for dimension reduction and food classification via harnessing multi-CNN model high-level featuresNEUROCOMPUTING
NEUROCOMPUTINGIF6.5
PREAI
分享收藏分享收藏
查看更多内容