Dynamic Weight-Enhanced Contrastive Learning for Partially View-Aligned Clustering

计算机科学 人工智能 聚类分析 模式识别(心理学) 特征(语言学) 特征提取 特征学习 模糊聚类 图像分割 自然语言处理 光谱聚类 对象(语法)
作者
Qiuyu Chen,Xihong Yang,Zhibing Dong,Qian Qu,En Zhu,Xinwang Liu
出处
期刊:IEEE Transactions on Circuits and Systems for Video Technology [Institute of Electrical and Electronics Engineers]
卷期号:36 (8): 12426-12438
标识
DOI:10.1109/tcsvt.2026.3679096
摘要

Deep contrastive multi-view clustering has attracted considerable attention for its ability to extract shared representations across heterogeneous views. Despite its effectiveness, existing approaches usually assume full cross-view alignment, an assumption often violated in practice due to sensor noise, transmission loss, or inconsistent data augmentation. This partial view alignment substantially degrades performance, while current solutions relying on fixed pair selection or uniform weighting fail to resolve the semantic ambiguity of unaligned samples. Consequently, latent cross-view semantic relationships remain underutilized, while erroneous positive pairs unduly dominate the contrastive objective, compromising robustness under low alignment rates or noisy conditions. To address these challenges, we propose Dynamic Weight-Enhanced InfoNCE (DW-InfoNCE), a contrastive learning framework designed for partially view-aligned clustering. The framework introduces a cross-view positive pair mining strategy that iteratively identifies reliable pairs from unaligned samples through quantile-adaptive thresholds and uniqueness constraints, enabling recovery of latent semantic structures obscured by incomplete alignment. In addition, an adaptive faulty pair suppression mechanism is developed to identify semantically inconsistent pairs via temporal loss smoothing and global statistics, while attenuating their impact with Gaussian-based adaptive weights. Extensive experiments on five benchmarks demonstrate DW-InfoNCE consistently outperforms state-of-the-art methods across alignment rates, with significant gains under severe misalignment. The code of DW-InfoNCE can be publicly downloaded at https://github.com/qiuyu99627/ DWInfoNCE.git.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
赘婿的应助被谨慎的秋烟采纳,获得10
刚刚
刚刚
Lucas的应助被程冬月采纳,获得10
1秒前
科研通AI6.2的应助被Arrebol采纳,获得10
2秒前
cherleen完成签到,获得积分10
2秒前
3秒前
lq发布了新的文献求助10
3秒前
3秒前
曹苍久发布了新的文献求助10
3秒前
4秒前
小困包发布了新的文献求助10
4秒前
4秒前
4秒前
movoandy发布了新的文献求助10
6秒前
6秒前
6秒前
思源的应助被梨花采纳,获得10
6秒前
7秒前
子非鱼发布了新的文献求助10
7秒前
9秒前
9秒前
zhangyu完成签到,获得积分10
9秒前
科研通AI6.4的应助被花卷卷采纳,获得10
9秒前
9秒前
科研通AI6.4的应助被花卷卷采纳,获得10
9秒前
阳光雨堂发布了新的文献求助10
9秒前
Eana发布了新的文献求助10
10秒前
科研通AI6.2的应助被152455采纳,获得10
10秒前
10秒前
科研通AI6.4的应助被152455采纳,获得30
10秒前
科研通AI6.2的应助被152455采纳,获得10
10秒前
美满不平发布了新的文献求助10
11秒前
科研通AI6.4的应助被152455采纳,获得10
11秒前
科研通AI6.4的应助被152455采纳,获得10
11秒前
科研通AI6.4的应助被152455采纳,获得10
11秒前
科研通AI6.4的应助被152455采纳,获得10
11秒前
秋风的应助被152455采纳,获得10
11秒前
SciGPT的应助被152455采纳,获得10
12秒前
陈博逊完成签到,获得积分20
12秒前
瓜洲发布了新的文献求助70
12秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
Markov Decision Processes and Reinforcement Learning 700
八旗制度の研究 600
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8171039
求助须知:如何正确求助?哪些是违规求助? 9563751
关于积分的说明 21772122
捐赠科研通 7762548
什么是DOI,文献DOI怎么找? 3430786
关于科研通互助平台的介绍 2554676
邀请新用户注册赠送积分活动 2475657