Search

Graph Neural Networks (GNNs) are powerful tools for handling graph-type data. Recently, GNNs have been widely applied in various domains, but they also face some issues, such as overfitting, over-smoothing and non-robustness. The existing research indicates that random dropout methods are an effective way to address these issues. However, random dropout methods in GNNs still face unresolved problems. Currently, the choice of dropout rate, often determined by heuristic or grid search methods, can increase the generalization error, contradicting the principal aims of dropout. In this paper, we propose a novel random dropout method for GNNs called FlexiDrop. First, we conduct a theoretical analysis of dropout in GNNs using rademacher complexity and demonstrate that the generalization error of traditional random dropout methods is constrained by a function related to the dropout rate. Subsequently, we use this function as a regularizer to unify the dropout rate and empirical loss within a single loss function, optimizing them simultaneously. Therefore, our method enables adaptive adjustment of the dropout rate and theoretically balances the trade-off between model complexity and generalization ability. Furthermore, extensive experimental results on benchmark datasets show that FlexiDrop outperforms traditional random dropout methods in GNNs.

Although 3D Gaussian Splatting (3DGS) has recently made progress in 3D human reconstruction, it primarily relies on 2D pixel-level supervision, overlooking the geometric complexity and topological relationships of different body parts. To address this gap, we introduce the Hierarchical Graph Human Gaussian Control (HUGS) framework for achieving high-fidelity 3D human reconstruction. Our approach involves leveraging explicitly semantic priors of body parts to ensure the consistency of geometric topology, thereby enabling the capture of the complex geometrical and topological associations among body parts. Additionally, we disentangle high-frequency features from global human features to refine surface details in body parts. Extensive experiments demonstrate that our method exhibits superior performance in human body reconstruction, particularly in enhancing surface details and accurately reconstructing body part junctions. Codes are available at https://wanghongsheng01.github.io/HUGS/.

We aimed to explore the capability of deep learning to approximate the function instantiated by biological neural circuits-the functional connectome. Using deep neural networks, we performed supervised learning with firing rate observations drawn from synthetically constructed neural circuits, as well as from an empirically supported Boundary Vector Cell-Place Cell network. The performance of trained networks was quantified using a range of criteria and tasks. Our results show that deep neural networks were able to capture the computations performed by synthetic biological networks with high accuracy, and were highly data efficient and robust to biological plasticity. We show that trained deep neural networks are able to perform zero-shot generalisation in novel environments, and allows for a wealth of tasks such as decoding the animal's location in space with high accuracy. Our study reveals a novel and promising direction in systems neuroscience, and can be expanded upon with a multitude of downstream applications, for example, goal-directed reinforcement learning., Comment: 13 pages, 10 figures

Yuemei Han,1,* Junmei Tang,1,* Jiayi Xia,1 Rui Wang,1 Chen Qin,1 Sihao Liu,1 Xia Zhao,1 Hao Chen,1,2 Quankui Lin1,2 1School of Ophthalmology & Optometry, Eye Hospital, Wenzhou Medical University, Wenzhou 325027, People’s Republic of China; 2Wenzhou Institute of Biomaterials and Engineering, Chinese Academy of Sciences, Wenzhou 32500, People’s Republic of China*These authors contributed equally to this workCorrespondence: Hao ChenSchool of Ophthalmology & Optometry, Eye Hospital, Wenzhou Medical University, 270 Xueyuan Xi Road, Wenzhou 325027, People’s Republic of ChinaTel/Fax +86 577 8806 7962Email chenhao823@mail.eye.ac.cnQuankui LinSchool of Ophthalmology & Optometry, Eye Hospital, Wenzhou Medical University, 270 Xueyuan Xi Road, Wenzhou 325027, People’s Republic of ChinaTel/Fax +86 577 8806 7962Email linqk@wmu.edu.cnBackground: Posterior capsular opacification (PCO) is the main complication after intraocular lens (IOL) implantation in cataract surgery, which is the result of lens epithelial cell (LEC) adhesion, proliferation and migration on the IOL and at the lens capsule interface. Hydrophilic surface modification, such as surface heparinization, decreases the cell adhesion, which has been commercialized and used clinically. However, clinical long-term observation results show no significant difference between the pristine and heparinized IOLs.Methods: To prevent PCO over the long time span, we modified the IOLs with an antiproliferative drug-loaded hydrophilic coating. The antiproliferative drug doxorubicin (DOX)-incorporated chitosan (CHI) nanoparticle was fabricated by sodium tripolyphosphate (TPP) gelation. Such antiproliferative drug-loaded CHI-TPP-DOX nanoparticles (CTDNP) were used as one of the building blocks to prepare polyelectrolyte multilayer with heparin (HEP) via layer-by-layer assembly, obtaining (HEP/CTDNP)n multilayers. The assembly process was characterized by quartz crystal micr

Yuemei Han,1,* Junmei Tang,1,* Jiayi Xia,1 Rui Wang,1 Chen Qin,1 Sihao Liu,1 Xia Zhao,1 Hao Chen,1,2 Quankui Lin1,2 1School of Ophthalmology & Optometry, Eye Hospital, Wenzhou Medical University, Wenzhou 325027, People’s Republic of China; 2Wenzhou Institute of Biomaterials and Engineering, Chinese Academy of Sciences, Wenzhou 32500, People’s Republic of China*These authors contributed equally to this workCorrespondence: Hao ChenSchool of Ophthalmology & Optometry, Eye Hospital, Wenzhou Medical University, 270 Xueyuan Xi Road, Wenzhou 325027, People’s Republic of ChinaTel/Fax +86 577 8806 7962Email chenhao823@mail.eye.ac.cnQuankui LinSchool of Ophthalmology & Optometry, Eye Hospital, Wenzhou Medical University, 270 Xueyuan Xi Road, Wenzhou 325027, People’s Republic of ChinaTel/Fax +86 577 8806 7962Email linqk@wmu.edu.cnBackground: Posterior capsular opacification (PCO) is the main complication after intraocular lens (IOL) implantation in cataract surgery, which is the result of lens epithelial cell (LEC) adhesion, proliferation and migration on the IOL and at the lens capsule interface. Hydrophilic surface modification, such as surface heparinization, decreases the cell adhesion, which has been commercialized and used clinically. However, clinical long-term observation results show no significant difference between the pristine and heparinized IOLs.Methods: To prevent PCO over the long time span, we modified the IOLs with an antiproliferative drug-loaded hydrophilic coating. The antiproliferative drug doxorubicin (DOX)-incorporated chitosan (CHI) nanoparticle was fabricated by sodium tripolyphosphate (TPP) gelation. Such antiproliferative drug-loaded CHI-TPP-DOX nanoparticles (CTDNP) were used as one of the building blocks to prepare polyelectrolyte multilayer with heparin (HEP) via layer-by-layer assembly, obtaining (HEP/CTDNP)n multilayers. The assembly process was characterized by quartz crystal micr