Contrastive learning has strong generalization ability and the capability to learn automatically without labeled information. However, it still faces challenges such as insufficient feature diversity, a lack of multi-level semantics, and the balance between tolerance and consistency. To address these challenges, This study propose a Multi-view Hierarchical Graph Contrastive Learning method. First, a new view is generated through a diffusion matrix to provide multi-view data for contrastive learning. Then, these multi-view data are fed into an asynchronous asymmetric network structure, specifically using graph network models to learn diversified features. Next, we adopt a self-designed hierarchical contrastive learning framework, constructing a three-level contrastive loss for joint optimization of nodes, subgraphs, and global graphs. Meanwhile, we introduce alignment and consistency and appropriately adjust the loss function through a temperature coefficient. Ultimately, the model achieves excellent classification performance on multiple datasets through node classification and graph classification tasks.

Emotion Recognition in Conversation (ERC) aims to detect the emotions of individual utterances within a conversation. Generating efficient and modality-specific representations for each utterance remains a significant challenge. Previous studies have proposed various models to integrate features extracted using different modality-specific encoders. However, they neglect the varying contributions of modalities to this task and introduce high complexity by aligning modalities at the frame level. To address these challenges, we propose the Multi-modal Anchor Gated Transformer with Knowledge Distillation (MAGTKD) for the ERC task. Specifically, prompt learning is employed to enhance textual modality representations, while knowledge distillation is utilized to strengthen representations of weaker modalities. Furthermore, we introduce a multi-modal anchor gated transformer to effectively integrate utterance-level representations across modalities. Extensive experiments on the IEMOCAP and MELD datasets demonstrate the effectiveness of knowledge distillation in enhancing modality representations and achieve state-of-the-art performance in emotion recognition. Our code is available at: https://github.com/JieLi-dd/MAGTKD.

Graph Neural Networks (GNNs) are powerful models for node classification, but their performance is heavily reliant on manually labeled data, which is often costly and results in insufficient labeling. Recent studies have shown that message-passing neural networks struggle to propagate information in low-degree nodes, negatively affecting overall performance. To address the information bias caused by degree imbalance, we propose a Learnable Enhancement and Label Selection Dynamic Graph Convolutional Network (L2DGCN). L2DGCN consists of a teacher model and a student model. The teacher model employs an improved label propagation mechanism that enables remote label information dissemination among all nodes. The student model introduces a dynamically learnable graph enhancement strategy, perturbing edges to facilitate information exchange among low-degree nodes. This approach maintains the global graph structure while learning graph representations. Additionally, we have designed a label selector to mitigate the impact of unreliable pseudo-labels on model learning. To validate the effectiveness of our proposed model with limited labeled data, we conducted comprehensive evaluations of semi-supervised node classification across various scenarios with a limited number of annotated nodes. Experimental results demonstrate that our data enhancement model significantly contributes to node classification tasks under sparse labeling conditions.

Since the regulatory roles of microRNAs in complex human diseases have been gradually demonstrated, more and more enlightening models were developed for predicting microRNA-disease associations. These models can inform bio-logical studies on the differential expression of microRNAs. In this paper, a graph convolutional network-based model using multi-view feature fusion and matrix completion, FMCGCN for brevity, is proposed. First of all, graph autoencoders are used to learn multiple embeddings from different networks. For convenience, attention mechanisms are used to fuse them. As a result, multi-view features constructed from different perspectives are aggregated. Then, the fused embedding is fed into the graph convolutional network to aggregate local information. This fused embedding is thought to facilitate feature extraction for graph convolutional networks. Finally, feature and nuclear norm minimization, a method for matrix completion, is used to obtain the prediction matrix. Additionally, evaluation results under 5-fold cross-validation prove that FMCGCN outperforms current models in several metrics. Furthermore, case studies for esophageal neoplasms and pancreatic neoplasms also demonstrate the validity of our model.

Convolutional neural networks (CNN) have achieved remarkable performance in image denoising. However, most existing CNNs cannot accurately capture and remove tiny noises during the denoising process and lose edge detail information easily. In this paper, we propose a fine-grained residual network guided by wavelet and adaptive coordinate attention (WACAFRN) for image denoising. Firstly, we propose an adaptive coordinate attention mechanism and combine it with cascaded Res2Net residual blocks to form an encoder network for more accurate noise removal. Secondly, we propose a wavelet attention mechanism that combines global and local residual blocks to form a decoder network, aiming to address the problem of edge detail information loss. At last, we complement the noise information through a noise estimation block to further enhance the model’s ability to adapt to noise. Extensive experiment results demonstrate that our proposed method outperforms existing denoising methods in both qualitative and quantitative aspects. Notably, our method significantly improves real-world noise removal tasks on the CC dataset, with an average increase of 2.08 dB in PSNR and 0.0264 in SSIM over the state-of-the-art methods. Additionally, WACAFRN exhibits faster inference speeds, underscoring its efficiency in real-world applications.

Collaborative filtering (CF) has proved to be effective for predicting users’ preferences. Embedding-based CF models have been the most popular ones for over a decade on very large datasets, among which the basic matrix factorization (MF) is the simplest but effective model. Latent factor models can capture global structure of the rating matrix but usually ignore local structure, which can be easily captured by neighborhood-based CF methods. Thus, a MF-based model that captures neighbor-entity spatial relationships between entities (users or items) should get better performance. Inspired by the recently developed capsule network which can capture part-whole spatial relationships, we incorporate capsules into CF to capture both local and global structure of the matrix. Specifically, we adopt the same way as MF to model the user-item interaction to capture global structure. To directly capture local structure in a similar way to neighborhood-based CF methods, we devise two types of routing mechanisms between capsules: pairwise routing and linear routing. These two types of routing algorithms can capture different levels of neighborhood relationships between entities and can be combined to further improve the performance of our models. Experimental results on Douban, MovieLens and Netflix datasets demonstrate that the proposed model achieves state-of-the-art performance.