Entropy-informed weighting channel normalizing flow for deep generative models

IF 7.6 1区 计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE
Wei Chen , Shian Du , Shigui Li , Delu Zeng , John Paisley
{"title":"Entropy-informed weighting channel normalizing flow for deep generative models","authors":"Wei Chen ,&nbsp;Shian Du ,&nbsp;Shigui Li ,&nbsp;Delu Zeng ,&nbsp;John Paisley","doi":"10.1016/j.patcog.2025.112442","DOIUrl":null,"url":null,"abstract":"<div><div>Normalizing Flows (NFs) are widely used in deep generative models for their exact likelihood estimation and efficient sampling. However, they require substantial memory since the latent space matches the input dimension. Multi-scale architectures address this by progressively reducing latent dimensions while preserving reversibility. Existing multi-scale architectures use simple, static channel-wise splitting, limiting expressiveness. To improve this, we introduce a regularized, feature-dependent <span><math><mi>Shuffle</mi></math></span> operation and integrate it into vanilla multi-scale architecture. This operation adaptively generates channel-wise weights and shuffles latent variables before splitting them. We observe that such operation guides the variables to evolve in the direction of entropy increase, hence we refer to NFs with the <span><math><mi>Shuffle</mi></math></span> operation as <em>Entropy-Informed Weighting Channel Normalizing Flow</em> (EIW-Flow). Extensive experiments on CIFAR-10, CelebA, ImageNet, and LSUN demonstrate that EIW-Flow achieves state-of-the-art density estimation and competitive sample quality for deep generative modeling, with minimal computational overhead.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"172 ","pages":"Article 112442"},"PeriodicalIF":7.6000,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Pattern Recognition","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0031320325011045","RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0

Abstract

Normalizing Flows (NFs) are widely used in deep generative models for their exact likelihood estimation and efficient sampling. However, they require substantial memory since the latent space matches the input dimension. Multi-scale architectures address this by progressively reducing latent dimensions while preserving reversibility. Existing multi-scale architectures use simple, static channel-wise splitting, limiting expressiveness. To improve this, we introduce a regularized, feature-dependent Shuffle operation and integrate it into vanilla multi-scale architecture. This operation adaptively generates channel-wise weights and shuffles latent variables before splitting them. We observe that such operation guides the variables to evolve in the direction of entropy increase, hence we refer to NFs with the Shuffle operation as Entropy-Informed Weighting Channel Normalizing Flow (EIW-Flow). Extensive experiments on CIFAR-10, CelebA, ImageNet, and LSUN demonstrate that EIW-Flow achieves state-of-the-art density estimation and competitive sample quality for deep generative modeling, with minimal computational overhead.
深度生成模型的熵通知加权信道归一化流
归一化流因其精确的似然估计和高效的采样而被广泛应用于深度生成模型中。然而,由于潜在空间与输入维度匹配,它们需要大量的内存。多尺度架构通过在保持可逆性的同时逐步减少潜在维度来解决这个问题。现有的多尺度架构使用简单的静态通道划分,限制了表现力。为了改进这一点,我们引入了一个正则化的、特征依赖的Shuffle操作,并将其集成到vanilla多尺度架构中。该操作自适应地生成通道权重,并在拆分潜在变量之前对其进行洗牌。我们观察到这种操作引导变量向熵增加的方向演化,因此我们将Shuffle操作的NFs称为熵知情加权通道归一化流(EIW-Flow)。在CIFAR-10、CelebA、ImageNet和LSUN上进行的大量实验表明,EIW-Flow以最小的计算开销实现了深度生成建模的最先进的密度估计和具有竞争力的样本质量。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Pattern Recognition
Pattern Recognition 工程技术-工程:电子与电气
CiteScore
14.40
自引率
16.20%
发文量
683
审稿时长
5.6 months
期刊介绍: The field of Pattern Recognition is both mature and rapidly evolving, playing a crucial role in various related fields such as computer vision, image processing, text analysis, and neural networks. It closely intersects with machine learning and is being applied in emerging areas like biometrics, bioinformatics, multimedia data analysis, and data science. The journal Pattern Recognition, established half a century ago during the early days of computer science, has since grown significantly in scope and influence.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信