LIVIVO - Search results -

Search results

Result 1 - 10 of total 87

Search options

Book ; Online: Generating Multi-Center Classifier via Conditional Gaussian Distribution

Zhang, Zhemin / Gong, Xun

2024

Abstract: The linear classifier is widely used in various image classification tasks. It works by optimizing the distance between a sample and its corresponding class center. However, in real-world data, one class can contain several local clusters, e.g., birds of ...

Abstract	The linear classifier is widely used in various image classification tasks. It works by optimizing the distance between a sample and its corresponding class center. However, in real-world data, one class can contain several local clusters, e.g., birds of different poses. To address this complexity, we propose a novel multi-center classifier. Different from the vanilla linear classifier, our proposal is established on the assumption that the deep features of the training set follow a Gaussian Mixture distribution. Specifically, we create a conditional Gaussian distribution for each class and then sample multiple sub-centers from that distribution to extend the linear classifier. This approach allows the model to capture intra-class local structures more efficiently. In addition, at test time we set the mean of the conditional Gaussian distribution as the class center of the linear classifier and follow the vanilla linear classifier outputs, thus requiring no additional parameters or computational overhead. Extensive experiments on image classification show that the proposed multi-center classifier is a powerful alternative to widely used linear classifiers. Code available at https://github.com/ZheminZhang1/MultiCenter-Classifier.
Keywords	Computer Science - Computer Vision and Pattern Recognition
Subject code	519
Publishing date	2024-01-29
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Book ; Online: Vision Big Bird

Zhang, Zhemin / Gong, Xun

Random Sparsification for Full Attention

2023

Abstract: Recently, Transformers have shown promising performance in various vision tasks. However, the high costs of global self-attention remain challenging for Transformers, especially for high-resolution vision tasks. Inspired by one of the most successful ... ...

Abstract	Recently, Transformers have shown promising performance in various vision tasks. However, the high costs of global self-attention remain challenging for Transformers, especially for high-resolution vision tasks. Inspired by one of the most successful transformers-based models for NLP: Big Bird, we propose a novel sparse attention mechanism for Vision Transformers (ViT). Specifically, we separate the heads into three groups, the first group used convolutional neural network (CNN) to extract local features and provide positional information for the model, the second group used Random Sampling Windows (RS-Win) for sparse self-attention calculation, and the third group reduces the resolution of the keys and values by average pooling for global attention. Based on these components, ViT maintains the sparsity of self-attention while maintaining the merits of Big Bird (i.e., the model is a universal approximator of sequence functions and is Turing complete). Moreover, our results show that the positional encoding, a crucial component in ViTs, can be safely removed in our model. Experiments show that Vision Big Bird demonstrates competitive performance on common vision tasks. Comment: arXiv admin note: substantial text overlap with arXiv:2304.06250
Keywords	Computer Science - Computer Vision and Pattern Recognition
Subject code	006
Publishing date	2023-11-10
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Book ; Online: RSIR Transformer

Zhang, Zhemin / Gong, Xun

Hierarchical Vision Transformer using Random Sampling Windows and Important Region Windows

2023

Abstract	Recently, Transformers have shown promising performance in various vision tasks. However, the high costs of global self-attention remain challenging for Transformers, especially for high-resolution vision tasks. Local self-attention runs attention computation within a limited region for the sake of efficiency, resulting in insufficient context modeling as their receptive fields are small. In this work, we introduce two new attention modules to enhance the global modeling capability of the hierarchical vision transformer, namely, random sampling windows (RS-Win) and important region windows (IR-Win). Specifically, RS-Win sample random image patches to compose the window, following a uniform distribution, i.e., the patches in RS-Win can come from any position in the image. IR-Win composes the window according to the weights of the image patches in the attention map. Notably, RS-Win is able to capture global information throughout the entire model, even in earlier, high-resolution stages. IR-Win enables the self-attention module to focus on important regions of the image and capture more informative features. Incorporated with these designs, RSIR-Win Transformer demonstrates competitive performance on common vision tasks. Comment: arXiv admin note: substantial text overlap with arXiv:2209.08726
Keywords	Computer Science - Computer Vision and Pattern Recognition
Subject code	600
Publishing date	2023-04-13
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Article ; Online: Structural basis of DNA recognition of the

Zhang, Zhemin / Yan, Yuqi / Pang, Jinji / Dai, Lei / Zhang, Qijing / Yu, Edward W

mBio

2024 Volume 15, Issue 3, Page(s) e0343023

Abstract: Campylobacter jejuni: Importance: Campylobacter ... ...

Abstract	Campylobacter jejuni Importance: Campylobacter jejuni
MeSH term(s)	Animals ; Humans ; Campylobacter jejuni/genetics ; Cryoelectron Microscopy ; Campylobacter/genetics ; Anti-Bacterial Agents/metabolism ; DNA/metabolism ; Bacterial Proteins/metabolism
Chemical Substances	Anti-Bacterial Agents ; DNA (9007-49-2) ; Bacterial Proteins
Language	English
Publishing date	2024-02-07
Publishing country	United States
Document type	Journal Article
ZDB-ID	2557172-2
ISSN	2150-7511 ; 2161-2129
ISSN (online)	2150-7511
ISSN	2161-2129
DOI	10.1128/mbio.03430-23
Database	MEDical Literature Analysis and Retrieval System OnLINE

Order via subito

This service is chargeable due to the Delivery terms set by subito. Orders including an article and supplementary material will be classified as separate orders. In these cases, fees will be demanded for each order.

Book ; Online: Positional Label for Self-Supervised Vision Transformer

Zhang, Zhemin / Gong, Xun

2022

Abstract: Positional encoding is important for vision transformer (ViT) to capture the spatial structure of the input image. General effectiveness has been proven in ViT. In our work we propose to train ViT to recognize the positional label of patches of the input ...

Abstract	Positional encoding is important for vision transformer (ViT) to capture the spatial structure of the input image. General effectiveness has been proven in ViT. In our work we propose to train ViT to recognize the positional label of patches of the input image, this apparently simple task actually yields a meaningful self-supervisory task. Based on previous work on ViT positional encoding, we propose two positional labels dedicated to 2D images including absolute position and relative position. Our positional labels can be easily plugged into various current ViT variants. It can work in two ways: (a) As an auxiliary training target for vanilla ViT (e.g., ViT-B and Swin-B) for better performance. (b) Combine the self-supervised ViT (e.g., MAE) to provide a more powerful self-supervised signal for semantic feature learning. Experiments demonstrate that with the proposed self-supervised methods, ViT-B and Swin-B gain improvements of 1.20% (top-1 Acc) and 0.74% (top-1 Acc) on ImageNet, respectively, and 6.15% and 1.14% improvement on Mini-ImageNet.
Keywords	Computer Science - Computer Vision and Pattern Recognition
Subject code	006
Publishing date	2022-06-10
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Book ; Online: The Fixed Sub-Center

Zhang, Zhemin / Gong, Xun

A Better Way to Capture Data Complexity

2022

Abstract: Treating class with a single center may hardly capture data distribution complexities. Using multiple sub-centers is an alternative way to address this problem. However, highly correlated sub-classes, the classifier's parameters grow linearly with the ... ...

Abstract	Treating class with a single center may hardly capture data distribution complexities. Using multiple sub-centers is an alternative way to address this problem. However, highly correlated sub-classes, the classifier's parameters grow linearly with the number of classes, and lack of intra-class compactness are three typical issues that need to be addressed in existing multi-subclass methods. To this end, we propose to use Fixed Sub-Center (F-SC), which allows the model to create more discrepant sub-centers while saving memory and cutting computational costs considerably. The F-SC specifically, first samples a class center Ui for each class from a uniform distribution, and then generates a normal distribution for each class, where the mean is equal to Ui. Finally, the sub-centers are sampled based on the normal distribution corresponding to each class, and the sub-centers are fixed during the training process avoiding the overhead of gradient calculation. Moreover, F-SC penalizes the Euclidean distance between the samples and their corresponding sub-centers, it helps remain intra-compactness. The experimental results show that F-SC significantly improves the accuracy of both image classification and fine-grained recognition tasks.
Keywords	Computer Science - Computer Vision and Pattern Recognition
Subject code	519
Publishing date	2022-03-24
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Article ; Online: Correlation Analysis and Prediction Model of Thermal Protection Performance of Aramid 1414 Fabric.

Zhang, Zhemin / Zhang, Jinzhong / Ye, Xiangyu / Ma, Keai / Li, Haihang

Polymers

2023 Volume 15, Issue 5

Abstract: The thermal protection performance of fire suit is vital to the safety of firefighters. Using certain physical properties of fabrics to evaluate their thermal protection performance speeds up the process. This work aims to develop a TPP value prediction ... ...

Abstract	The thermal protection performance of fire suit is vital to the safety of firefighters. Using certain physical properties of fabrics to evaluate their thermal protection performance speeds up the process. This work aims to develop a TPP value prediction model that can be easily applied. Five properties of three types of Aramid 1414 made of the same material were tested, and the relationships between the physical properties of Aramid 1414 and its thermal protection performance (TPP value) were investigated. The results showed that the TPP value of the fabric had a positive correlation with grammage and air gap, and a negative correlation with the underfill factor. A stepwise regression analysis was used to solve the collinearity issue between the independent variables. Finally, a model for predicting TPP value by air gap and underfill factor was developed. The method adopted in this work reduced the number of independent variables in the prediction model, which is conducive to the application of the model.
Language	English
Publishing date	2023-02-27
Publishing country	Switzerland
Document type	Journal Article
ZDB-ID	2527146-5
ISSN	2073-4360 ; 2073-4360
ISSN (online)	2073-4360
ISSN	2073-4360
DOI	10.3390/polym15051188
Database	MEDical Literature Analysis and Retrieval System OnLINE

Order via subito

Article ; Online: Cryo-EM Structures of AcrD Illuminate a Mechanism for Capturing Aminoglycosides from Its Central Cavity.

Zhang, Zhemin / Morgan, Christopher E / Cui, Meng / Yu, Edward W

mBio

2023 Volume 14, Issue 1, Page(s) e0338322

Abstract: The Escherichia coli acriflavine resistance protein D (AcrD) is an efflux pump that belongs to the resistance-nodulation-cell division (RND) superfamily. Its primary function is to provide resistance to aminoglycoside-based drugs by actively extruding ... ...

Abstract	The Escherichia coli acriflavine resistance protein D (AcrD) is an efflux pump that belongs to the resistance-nodulation-cell division (RND) superfamily. Its primary function is to provide resistance to aminoglycoside-based drugs by actively extruding these noxious compounds out of E. coli cells. AcrD can also mediate resistance to a limited range of other amphiphilic agents, including bile acids, novobiocin, and fusidic acids. As there is no structural information available for any aminoglycoside-specific RND pump, here we describe cryo-electron microscopy (cryo-EM) structures of AcrD in the absence and presence of bound gentamicin. These structures provide new information about the RND superfamily of efflux pumps, specifically, that three negatively charged residues central to the aminoglycoside-binding site are located within the ceiling of the central cavity of the AcrD trimer. Thus, it is likely that AcrD is capable of picking up aminoglycosides via this central cavity. Through the combination of cryo-EM structural determination, mutagenesis analysis, and molecular simulation, we show that charged residues are critically important for this pump to shuttle drugs directly from the central cavity to the funnel of the AcrD trimer for extrusion.
MeSH term(s)	Aminoglycosides ; Membrane Transport Proteins/metabolism ; Escherichia coli/genetics ; Escherichia coli/metabolism ; Cryoelectron Microscopy ; Acriflavine/metabolism ; Anti-Bacterial Agents/pharmacology ; Gentamicins ; Drug Resistance, Multiple, Bacterial
Chemical Substances	Aminoglycosides ; Membrane Transport Proteins ; Acriflavine (1T3A50395T) ; Anti-Bacterial Agents ; Gentamicins
Language	English
Publishing date	2023-01-10
Publishing country	United States
Document type	Journal Article ; Research Support, N.I.H., Extramural
ZDB-ID	2557172-2
ISSN	2150-7511 ; 2161-2129
ISSN (online)	2150-7511
ISSN	2161-2129
DOI	10.1128/mbio.03383-22
Database	MEDical Literature Analysis and Retrieval System OnLINE

Order via subito

Article ; Online: Cryo-EM Structures of the Klebsiella pneumoniae AcrB Multidrug Efflux Pump.

Zhang, Zhemin / Morgan, Christopher E / Bonomo, Robert A / Yu, Edward W

mBio

2023 Volume 14, Issue 3, Page(s) e0065923

Abstract: The continued challenges of the COVID-19 pandemic combined with the growing problem of antimicrobial-resistant bacterial infections has severely impacted global health. Specifically, the Gram-negative pathogen Klebsiella pneumoniae is one of the most ... ...

Abstract	The continued challenges of the COVID-19 pandemic combined with the growing problem of antimicrobial-resistant bacterial infections has severely impacted global health. Specifically, the Gram-negative pathogen Klebsiella pneumoniae is one of the most prevalent causes of secondary bacterial infection in COVID-19 patients, with approximately an 83% mortality rate observed among COVID-19 patients with these bacterial coinfections. K. pneumoniae belongs to the ESKAPE group of pathogens, a group that commonly gives rise to severe infections that are often life-threatening. Recently, K. pneumoniae carbapenemase (KPC)-producing K. pneumoniae has drawn wide public attention, as the mortality rate for this infection can be as high as 71%. The most predominant and clinically important multidrug efflux system in K. pneumoniae is the acriflavine resistance B (AcrB) multidrug efflux pump. This pump mediates resistance to different classes of structurally diverse antimicrobial agents, including quinolones, β-lactams, tetracyclines, macrolides, aminoglycosides, and chloramphenicol. We here report single-particle cryo-electron microscopy (cryo-EM) structures of K. pneumoniae AcrB, in both the absence and the presence of the antibiotic erythromycin. These structures allow us to elucidate specific pump-drug interactions and pinpoint exactly how this pump recognizes antibiotics.
MeSH term(s)	Humans ; Acriflavine/pharmacology ; Klebsiella pneumoniae ; Cryoelectron Microscopy ; Pandemics ; COVID-19 ; Anti-Bacterial Agents/pharmacology ; Bacterial Proteins/pharmacology ; Erythromycin ; Klebsiella Infections/microbiology ; Drug Resistance, Multiple, Bacterial ; Microbial Sensitivity Tests
Chemical Substances	Acriflavine (1T3A50395T) ; Anti-Bacterial Agents ; Bacterial Proteins ; Erythromycin (63937KV33D)
Language	English
Publishing date	2023-04-17
Publishing country	United States
Document type	Journal Article ; Research Support, N.I.H., Extramural ; Research Support, U.S. Gov't, Non-P.H.S.
ZDB-ID	2557172-2
ISSN	2150-7511 ; 2161-2129
ISSN (online)	2150-7511
ISSN	2161-2129
DOI	10.1128/mbio.00659-23
Database	MEDical Literature Analysis and Retrieval System OnLINE

Order via subito

Article ; Online: sRNA21, a novel small RNA, protects Mycobacterium abscessus against oxidative stress.

Tan, Zhili / Fan, Junsheng / He, Siyuan / Zhang, Zhemin / Chu, Haiqing

The journal of gene medicine

2023 Volume 25, Issue 6, Page(s) e3492

Abstract: Background: During infection, Mycobacterium abscessus encounters numerous environmental changes and adapts to them using a variety of complex mechanisms. Non-coding small RNAs (sRNAs) have been shown in other bacteria to be involved in post- ... ...

Abstract	Background: During infection, Mycobacterium abscessus encounters numerous environmental changes and adapts to them using a variety of complex mechanisms. Non-coding small RNAs (sRNAs) have been shown in other bacteria to be involved in post-transcriptional regulatory pathways, including environmental stress adaptation. However, the potential role of sRNAs in the resistance to oxidative stress in M. abscessus was not clearly described. Methods: In the present study, we analyzed putative sRNAs identified by RNA-sequencing (RNA-seq) experiments in M. abscessus ATCC_19977 under oxidative stress, and the transcription profiles of sRNAs with differential expression were verified by quantitative reverse transcription-PCR (qRT-PCR). Six sRNA overexpression strains were constructed, and the differences in growth curves between these strains and the control strain were verified. An upregulated sRNA under oxidative stress was selected and named sRNA21. The survival ability of the sRNA21 overexpression strain was assessed, and computer-based approaches were used to predict the targets and pathways regulated by sRNA21. The total ATP production and NAD Results: In total, 14 putative sRNAs were identified under oxidative stress, and the qRT-PCR analysis of six sRNAs showed comparable results to RNA-seq assays. Overexpression of sRNA21 in M. abscessus increased cell growth rate and intracellular ATP level before and after peroxide exposure. The expression of genes encoding alkyl hydroperoxidase and superoxide dismutase was significantly increased, and superoxide dismutase activity was enhanced in the sRNA21 overexpression strain. Meanwhile, after sRNA21 overexpression, the intracellular NAD Conclusions: Our findings show that sRNA21 is an oxidative stress-induced sRNA that increases M. abscessus survival and promotes the expression of antioxidant enzymes under oxidative stress. These findings may provide new insights into the adaptive transcriptional response of M. abscessus to oxidative stress.
MeSH term(s)	Mycobacterium abscessus/genetics ; Mycobacterium abscessus/metabolism ; NAD/metabolism ; RNA, Bacterial/genetics ; RNA, Bacterial/metabolism ; RNA, Small Untranslated/genetics ; RNA, Small Untranslated/metabolism ; Oxidative Stress/genetics ; Adenosine Triphosphate/metabolism
Chemical Substances	NAD (0U46U6E8UK) ; RNA, Bacterial ; RNA, Small Untranslated ; Adenosine Triphosphate (8L70Q75FXE)
Language	English
Publishing date	2023-03-18
Publishing country	England
Document type	Journal Article ; Research Support, Non-U.S. Gov't
ZDB-ID	1458024-x
ISSN	1521-2254 ; 1099-498X
ISSN (online)	1521-2254
ISSN	1099-498X
DOI	10.1002/jgm.3492
Database	MEDical Literature Analysis and Retrieval System OnLINE

Full text online

Accessible to users with ZB MED library card

In stock of ZB MED Cologne/Königswinter

Zs.A 5423: Show issues

Location:
Je nach Verfügbarkeit (siehe Angabe bei Bestand)
bis Jg. 1994: Bestellungen von Artikeln über das Online-Bestellformular
Jg. 1995 - 2021: Lesesall (2.OG)
ab Jg. 2022: Lesesaal (EG)

Order via subito

Details ▾

To top

Full text online

More links

Kategorien

Inter-library loan at ZB MED

Full text online

More links

Kategorien

Inter-library loan at ZB MED

Full text online

More links

Kategorien

Inter-library loan at ZB MED

More links

Kategorien

Order via subito

Full text online

More links

Kategorien

Inter-library loan at ZB MED

Full text online

More links

Kategorien

Inter-library loan at ZB MED

More links

Kategorien

Order via subito

More links

Kategorien

Order via subito

More links

Kategorien

Order via subito

Full text online

More links

Kategorien

In stock of ZB MED Cologne/Königswinter

Order via subito