Search

Your search keyword '"Guo, Yurong"' showing total 694 results
Start Over Author "Guo, Yurong"
694 results on '"Guo, Yurong"'

1. Focus the Overlapping Problem on Few-Shot Object Detection via Multiple Predictions

Author

Guan, Mandan, Yu, Wenqing, Guo, Yurong, Huang, Keyan, Zhang, Jiaxun, Liang, Kongming, Ma, Zhanyu, Goos, Gerhard, Founding Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Liu, Qingshan, editor, Wang, Hanzi, editor, Ma, Zhanyu, editor, Zheng, Weishi, editor, Zha, Hongbin, editor, Chen, Xilin, editor, Wang, Liang, editor, and Ji, Rongrong, editor

Published
2024
Full Text
View/download PDF

14. TLRM: Task-level Relation Module for GNN-based Few-Shot Learning

Author

Guo, Yurong, Ma, Zhanyu, Li, Xiaoxu, and Dong, Yuan

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Recently, graph neural networks (GNNs) have shown powerful ability to handle few-shot classification problem, which aims at classifying unseen samples when trained with limited labeled samples per class. GNN-based few-shot learning architectures mostly replace traditional metric with a learnable GNN. In the GNN, the nodes are set as the samples embedding, and the relationship between two connected nodes can be obtained by a network, the input of which is the difference of their embedding features. We consider this method of measuring relation of samples only models the sample-to-sample relation, while neglects the specificity of different tasks. That is, this method of measuring relation does not take the task-level information into account. To this end, we propose a new relation measure method, namely the task-level relation module (TLRM), to explicitly model the task-level relation of one sample to all the others. The proposed module captures the relation representations between nodes by considering the sample-to-task instead of sample-to-sample embedding features. We conducted extensive experiments on four benchmark datasets: mini-ImageNet, tiered-ImageNet, CUB-$200$-$2011$, and CIFAR-FS. Experimental results demonstrate that the proposed module is effective for GNN-based few-shot learning., Comment: VCIP 2021

Published
2021

16. Competing Ratio Loss for Discriminative Multi-class Image Classification

Author

Zhang, Ke, Guo, Yurong, Wang, Xinsheng, Chang, Dongliang, Zhao, Zhenbing, Ma, Zhanyu, and Han, Tony X.

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

The development of deep convolutional neural network architecture is critical to the improvement of image classification task performance. Many image classification studies use deep convolutional neural network and focus on modifying the network structure to improve image classification performance. Conversely, our study focuses on loss function design. Cross-entropy Loss (CEL) has been widely used for training deep convolutional neural network for the task of multi-class classification. Although CEL has been successfully implemented in several image classification tasks, it only focuses on the posterior probability of the correct class. For this reason, a negative log likelihood ratio loss (NLLR) was proposed to better differentiate between the correct class and the competing incorrect ones. However, during the training of the deep convolutional neural network, the value of NLLR is not always positive or negative, which severely affects the convergence of NLLR. Our proposed competing ratio loss (CRL) calculates the posterior probability ratio between the correct class and the competing incorrect classes to further enlarge the probability difference between the correct and incorrect classes. We added hyperparameters to CRL, thereby ensuring its value to be positive and that the update size of backpropagation is suitable for the CRL's fast convergence. To demonstrate the performance of CRL, we conducted experiments on general image classification tasks (CIFAR10/100, SVHN, ImageNet), the fine-grained image classification tasks (CUB200-2011 and Stanford Car), and the challenging face age estimation task (using Adience). Experimental results show the effectiveness and robustness of the proposed loss function on different deep convolutional neural network architectures and different image classification tasks., Comment: arXiv admin note: substantial text overlap with arXiv:1907.13349

Published
2019

39. The gene product of a Trypanosoma equiperdum ortholog of the cAMP-dependent protein kinase regulatory subunit is a monomeric protein that is not capable of binding cyclic nucleotides

Author

Bubis, José, Martínez, Juan Carlos, Calabokis, Maritza, Ferreira, Joilyneth, Sanz-Rodríguez, Carlos E, Navas, Victoria, Escalona, José Leonardo, Guo, Yurong, and Taylor, Susan S

Subjects
Biochemistry and Cell Biology, Biological Sciences, Rare Diseases, Amino Acid Sequence, Base Sequence, Cyclic AMP-Dependent Protein Kinases, Molecular Docking Simulation, Protein Binding, Protein Conformation, Sequence Homology, Nucleic Acid, Trypanosoma, Trypanosoma equiperdum, Regulatory subunits of the cAMP-dependent protein kinase, Gene cloning, Protein expression in bacteria, Protein purification, Biochemical characterization, Dourine, Trypanosomatid parasites, Biochemistry & Molecular Biology, Biochemistry and cell biology
Abstract

The full gene sequence encoding for the Trypanosoma equiperdum ortholog of the cAMP-dependent protein kinase (PKA) regulatory (R) subunits was cloned. A poly-His tagged construct was generated [TeqR-like(His)8], and the protein was expressed in bacteria and purified to homogeneity. The size of the purified TeqR-like(His)8 was determined to be ∼57,000 Da by molecular exclusion chromatography indicating that the parasite protein is a monomer. Limited proteolysis with various proteases showed that the T. equiperdum R-like protein possesses a hinge region very susceptible to proteolysis. The recombinant TeqR-like(His)8 did not bind either [3H] cAMP or [3H] cGMP up to concentrations of 0.40 and 0.65 μM, respectively, and neither the parasite protein nor its proteolytically generated carboxy-terminal large fragments were capable of binding to a cAMP-Sepharose affinity column. Bioinformatics analyses predicted that the carboxy-terminal region of the trypanosomal R-like protein appears to fold similarly to the analogous region of all known PKA R subunits. However, the protein amino-terminal portion seems to be unrelated and shows homology with proteins that contained Leu-rich repeats, a folding motif that is particularly appropriate for protein-protein interactions. In addition, the three-dimensional structure of the T. equiperdum protein was modeled using the crystal structure of the bovine PKA RIα subunit as template. Molecular docking experiments predicted critical changes in the environment of the two putative nucleotide binding clefts of the parasite protein, and the resulting binding energy differences support the lack of cyclic nucleotide binding in the trypanosomal R-like protein.

Published
2018

Catalog

Books, media, physical & digital resources
See catalog results