Results 101 to 110 of about 115,093 (312)

Contrasting the landscape of contrastive and non-contrastive learning

, 2022
A lot of recent advances in unsupervised feature learning are based on designing features which are invariant under semantic data augmentations. A common way to do this is contrastive learning, which uses positive and negative samples. Some recent works however have shown promising results for non-contrastive learning, which does not require negative ...
Pokle, Ashwini, Tian, Jinjin, Li, Yuchen, Risteski, Andrej   +3 more
openaire   +1 more source

Enzymatic degradation of biopolymers in amorphous and molten states: mechanisms and applications

FEBS Open Bio, EarlyView.
This review explains how polymer morphology and thermal state shape enzymatic degradation pathways, comparing amorphous and molten biopolymer structures. By integrating structure–reactivity principles with insights from thermodynamics and enzyme engineering, it highlights mechanisms that enable efficient polymer breakdown.
Anđela Pustak, Aleksandra Maršavelski
wiley   +1 more source

The Aging Blood: Cellular Origins, Circulating Drivers, and Therapeutic Potential

Aging and Cancer, EarlyView.
As a conduit linking all organs, the blood system both reflects and actively drives systemic aging. This review highlights how circulating pro‐aging and antiaging factors and age‐associated hematopoietic stem cell dysfunction contribute to immunosenescence and multi‐organ decline, positioning the hematopoietic system as a target for aging intervention.
Hanqing He, Jianwei Wang
wiley   +1 more source

A Theoretical Analysis of Contrastive Unsupervised Representation Learning

, 2019
Recent empirical works have successfully used unlabeled data to learn feature representations that are broadly useful in downstream classification tasks.
Arora, Sanjeev, Khandeparkar, Hrishikesh, Khodak, Mikhail, Plevrakis, Orestis, Saunshi, Nikunj   +4 more
core  

Edge Contrastive Learning: An Augmentation-Free Graph Contrastive Learning Model

Proceedings of the AAAI Conference on Artificial Intelligence
Graph contrastive learning (GCL) aims to learn representations from unlabeled graph data in a self-supervised manner and has developed rapidly in recent years. However, edge-level contrasts are not well explored by most existing GCL methods. Most studies in GCL only regard edges as auxiliary information while updating node features.
Li, Yujun, Zhang, Hongyuan, Yuan, Yuan
openaire   +2 more sources

Systemic T Cell Receptor Profiling Reveals Adaptive Immune Activation and Potential Immune Signatures of Diagnosis and Brain Atrophy in Epilepsy

Annals of Clinical and Translational Neurology, EarlyView.
ABSTRACT Objective Epilepsy is increasingly associated with immune dysregulation and inflammation. The T cell receptor (TCR), a key mediator of adaptive immunity, shows repertoire alterations in various immune‐mediated diseases. The unique TCR sequence serves as a molecular barcode for T cells, and clonal expansion accompanied by reduced overall TCR ...
Yong‐Won Shin, Sang Bin Hong, Yong Woo Shin, Inpyeong Hwang, Jaeseong Oh, Jihyeon Choi, Narae Kim, Jangsup Moon, Keun‐Hwa Jung, Kyung‐Il Park, Ki‐Young Jung, Kon Chu, Sang Kun Lee   +12 more
wiley   +1 more source

The Implementation of Contrastive Analysis-Based Arabic Learning

Alsinatuna, 2018
As well as other learning process, the successful Arabic learning can be achieved through several steps. This study aims to explain those steps, and it uses library research. Method of this study covers finding data and sources related to the contrastive
Raswan Raswan
doaj   +1 more source

Plasma Proteomic Signatures for Alzheimer's Disease: Comparable Accuracy to ATN Biomarkers and Cross‐Platform Validation

Annals of Clinical and Translational Neurology, EarlyView.
ABSTRACT Background There is growing recognition of the potential of plasma proteomics for Alzheimer's Disease (AD) risk assessment and disease characterization. However, differences between proteomics platforms introduce uncertainties regarding cross‐platform applicability.
Manyue Hu, Oliver Robinson, Christina M. Lill, Anna Matton, Raquel Puerta, Pilar Sanz, Merce Boada, Agustín Ruiz, Lefkos Middleton, for the Alzheimer's Disease Neuroimaging Initiative   +9 more
wiley   +1 more source

Higher Amyloid and Tau Burden Is Associated With Faster Decline on a Digital Cognitive Test

Annals of Clinical and Translational Neurology, EarlyView.
ABSTRACT Objective A 2‐min digital clock‐drawing test (DCTclock) captures more granular features of the clock‐drawing process than the pencil‐and‐paper clock‐drawing test, revealing more subtle deficits at the preclinical stage of Alzheimer's disease (AD). A previous cross‐sectional study demonstrated that worse DCTclock performance was associated with
Jessie Fanglu Fu, Talia Robinson, Marina Rodriguez Alonso, Adela Francis Malave, Roos J. Jutten, Michael J. Properzi, Brian C. Healy, Jackson C. Thompson, Grace Del Carmen Montenegro, Emma Thibault, Dana Penney, Randall Davis, Reisa A. Sperling, Kathryn V. Papp, Keith A. Johnson, Julie C. Price, Dorene M. Rentz   +16 more
wiley   +1 more source

Prototypical Graph Contrastive Learning for Recommendation

Applied Sciences
Data sparsity caused by limited interactions makes it challenging for recommendation to accurately capture user preferences. Contrastive learning effectively alleviates this issue by enriching embedding information through the learning of diverse ...
Tao Wei, Changchun Yang, Yanqi Zheng
doaj   +1 more source