Dezhong Peng

Divide and Conquer: Region Self-Paced Physics-Informed Neural Networks

Ruixuan Meng, Wenyuan Wu, Zhenwen Ren, Dezhong Peng et autres

Recently, Physics-Informed Neural Networks (PINNs) have shown strong potential in scientific computing and engineering simulation. Although existing PINN methods have achieved promising progress, they still suffer from the Unbalanced Prediction Problem (UPP), where challenging regions are learned much worse than smooth regions. …

cn (code pays fourni par la source)

NOTO: Noise-Tolerate Evidential Learning for Open-Set Cross-Modal Retrieval

Ruitao Pu, Chao Su, Peng Hu, Zhenwen Ren et autres

With the increasing accessibility of multimodal data, cross-modal retrieval (CMR) has gained significant attention in recent years. However, most existing CMR methods are built on clean annotations and closed-set label space assumptions, which are often violated in practice. In realistic scenarios, annotations …

cn (code pays fourni par la source)

Learning Fully Unsupervised Cross-domain Image Retrieval through Self-supervised Pseudo-labels

Yongxiang Li, Xiaoyun Ren, Yuan Sun, Guiduo Duan et autres

cn (code pays fourni par la source)

GBNOD: Granular-ball neighborhood outlier detection

Xinyu Su, Zheng Li, Dezhong Peng, Hongmei Chen

cn (code pays fourni par la source)

Semantic-Consistent Bidirectional Contrastive Hashing for Noisy Multi-Label Cross-Modal Retrieval

Chunhua Su, Wenyuan Wu, Yuan Sun, Dezhong Peng et autres

Cross-modal hashing (CMH) facilitates efficient retrieval across different modalities (e.g., image and text) by encoding data into compact binary representations. While recent methods have achieved remarkable performance, they often rely heavily on fully annotated datasets, which are costly and labor-intensive to obtain. …

Noise-Robust Cross-modal Learning for Reliable 2D-3D Retrieval

Yanglin Feng, Yuan Sun, Dezhong Peng, Guiduo Duan et autres

With the rapid proliferation of 2D and 3D data, driven by advances in virtual environments and AI-generated content, cross-modal 2D-3D retrieval has attracted growing attention. However, it is easy to introduce noisy labels due to the spatial complexity of 3D content. Although …

cn (code pays fourni par la source)

Multi-view Hashing Classification

Shilin Xu, Chao Su, Run Ye, Dezhong Peng et autres

Multi-view classification aims to leverage information from multiple views of data to improve prediction performance by learning complementary and consistent representations. Therefore, in recent years, multi-view learning has attracted widespread attention in the community. Despite the success of existing multi-view learning methods, …

cn (code pays fourni par la source)

Fuzzy $k$kNN Entropy and its Anomaly Detection

Zhong Yuan, Yuan Yuan, Dezhong Peng, Xiaomin Song

With the successful application of granular computing in anomaly detection, a variety of tools including fuzzy information entropy can achieve superior detection results. However, fuzzy information entropy calculates fuzzy similarity through a global strategy, ignoring the local information in the data. To …

cn, it (code pays fourni par la source)

A kernelized fuzzy approximation fusion model with granular-ball computing for outlier detection

Yongxiang Li, Xinyu Su, Run Ye, Dezhong Peng et autres

cn (code pays fourni par la source)

Identifying Outliers via Local Granular-Ball Density

Xinyu Su, Xiwen Wang, Dezhong Peng, Xiaomin Song et autres

Existing density-based outlier detection methods process data at the single-granularity level of individual samples, requiring pairwise distance calculations between all samples and exhibiting high sensitivity to noise. The single-granularity-based processing paradigm fails to mine the information at multiple levels of granularity in …

cn (code pays fourni par la source)

ROLL: Robust Noisy Pseudo-label Learning for Multi-View Clustering with Noisy Correspondence

Yuan Sun, Yongxiang Li, Zhenwen Ren, Guiduo Duan et autres

Multi-view clustering (MVC) aims to exploit complementary information from diverse views to enhance clustering performance. Since pseudo-labels can provide additional semantic information, many MVC methods have been proposed to guide unsupervised multi-view learning through pseudo-labels. These methods implicitly assume that the predicted …

cn (code pays fourni par la source)

Neuroimaging studies of acupuncture for depressive disorder: a systematic review of published papers from 2014 to 2024

Dezhi Lin, Qiang Ren, Yangxu Ou, Longlong Li et autres

Background: Several neuroimaging studies have confirmed that acupuncture can elicit alterations in brain networks and regions associated with depressive disorder (DD). This review provides an overview of the methodologies and results of neuroimaging investigations into the efficacy of acupuncture in treating DD, …

cn (code pays fourni par la source)