第二种,full graph,将多种views两两互相匹配。很明显的,full graph的交互信息更多,效果也更好,副作用是运算量也大很多。 4、Connecting to Mutual Information: 重头戏来了。 其实这一系列基于contrastive learning范式的学习方法,都直接关系到对 z_{i}=f_{\theta i}\left(v_{i}\right) 和z_{j}=f_{\...
CMC是一种 contrastive 的表示学习,使用基于 memory-bank 的negative样本采样方法。一般而言,contrastive learning 可以学到 anchor 样本和 positive 样本共同的特征表示,同时拉开 anchor 和 negative 在特征空间的距离。本文提出,可以将 anchor 和 positive 这种一对一的关系扩展为一对多甚至是多对多的关系,即 multi-vi...
Here, we present a novel interpretable model for coupling the structure and function activity of brain based on heterogeneous contrastive graph representation. The proposed method is able to link manifest variables of the brain (i.e. MEG, MRI, fMRI and behavior performance) and quantify the ...
7. Deep learning based or network based methods Part A 11 self-supervised learning (or contrastive learning) is also based on Deep learning. 9. Co-training based methods 9.1 JMLR20 Self-paced Multi-view Co-training(python) 10.1 IJCAI18 FISH-MML: Fisher-HSIC Multi-View Metric Learning(matlab...
2023Multi-channel Augmented Graph Embedding Convolutional Network for Multi-view ClusteringMAGEC-NetTNSE- 2022Deep Safe Multi-View Clustering:Reducing the Risk of Clustering Performance Degradation Caused by View IncreaseDSMVCCVPR 2022Multi-level Feature Learning for Contrastive Multi-view ClusteringMFLVCCVPR...
In particular, we combined fine-grained intents with a knowledge graph to calculate intent weights and capture intent semantics. The IKMCL model performs multiview intent contrastive learning at both coarse-grained and fine-grained levels to extract semantic relationships in user–item interactions and...
NeurIPS: Multi-view Contrastive Graph Clustering(MCGC).[Paper] [Code] TKDE: Self-supervised Discriminative Feature Learning for Deep Multi-view Clustering(SDMVC).[Paper] [Code] TKDE: Multi-view Attributed Graph Clustering(MAGC).[Paper] [Code] TMM: Deep Multi-view Subspace Clustering with ...
2023Self-Weighted Contrastive Learning among Multiple Views for Mitigating Representation DegenerationSEMNeurIPS 2023A Novel Approach for Effective Multi-View Clustering with Information-Theoretic PerspectiveSUMVCNeurIPS 2023Dual Label-Guided Graph Refnement for Multi-View Graph ClusteringDuaLGRAAAI ...
CoST: Contrastive Learning of Disentangled Seasonal-Trend Representations for Time Series Forecasting. arXiv 2022, arXiv:2202.01575. [Google Scholar] Challu, C.; Olivares, K.G.; Oreshkin, B.N.; Garza, F.; Mergenthaler, M.; Dubrawski, A. N-hits: Neural hierarchical interpolation for time...