Results 61 to 70 of about 98,849 (283)

Network Binarization via Contrastive Learning

, 2022
Neural network binarization accelerates deep models by quantizing their weights and activations into 1-bit. However, there is still a huge performance gap between Binary Neural Networks (BNNs) and their full-precision (FP) counterparts. As the quantization error caused by weights binarization has been reduced in earlier works, the activations ...
Shang, Yuzhang, Xu, Dan, Zong, Ziliang, Nie, Liqiang, Yan, Yan   +4 more
openaire   +2 more sources

Dual targeting of RET and SRC synergizes in RET fusion‐positive cancer cells

Molecular Oncology, EarlyView.
Despite the strong activity of selective RET tyrosine kinase inhibitors (TKIs), resistance of RET fusion‐positive (RET+) lung cancer and thyroid cancer frequently occurs and is mainly driven by RET‐independent bypass mechanisms. Son et al. show that SRC TKIs significantly inhibit PAK and AKT survival signaling and enhance the efficacy of RET TKIs in ...
Juhyeon Son, Lily L. Remsing Rix, Bin Fang, Eric A. Welsh, Nicole V. Bremer, Valentina Foglizzo, Paola Roa, Nickole Sigcha‐Coello, Yi Liao, Eric B. Haura, Alexander Drilon, John M. Koomen, Emiliano Cocco, Uwe Rix   +13 more
wiley   +1 more source

Children’s learning from contrast modelling [PDF]

Cognitive Development, 2002
This study investigates the effectiveness of immediately modelling the correct solution to a task on which children were making errors. The technique is based on proposals by Saxton (1997) who, in his contrast theory of negative input, claims that corrective speech input is particularly effective when it immediately follows a child's error, such as an ...
Pine, K. J., Messer, D. J., St. John, K.
openaire   +1 more source

Crucial parameters for precise copy number variation detection in formalin‐fixed paraffin‐embedded solid cancer samples

Molecular Oncology, EarlyView.
This study shows that copy number variations (CNVs) can be reliably detected in formalin‐fixed paraffin‐embedded (FFPE) solid cancer samples using ultra‐low‐pass whole‐genome sequencing, provided that key (pre)‐analytical parameters are optimized.
Hanne Goris, Vasiliki Siozopoulou, Léon C van Kempen, Anne Sieben, Ella Roelant, Stig Hellemans, Elyne Backx, Laure Sorber, Koen De Winne, Senada Koljenović, Karen Zwaenepoel   +10 more
wiley   +1 more source

Node classification in complex networks based on multi-view debiased contrastive learning

Complex & Intelligent Systems
In complex networks, contrastive learning has emerged as a crucial technique for acquiring discriminative representations from graph data. Maximizing the similarity among relevant sample pairs while minimizing that among irrelevant pairs is pivotal in ...
Zhe Li, Lei Zhou, Yandong Hou, Min Ji, Zhuanzheng Hang, Bolun Chen   +5 more
doaj   +1 more source

Dual Contrastive Learning Model Based Background Debiasing in SAR ATR

Shuju Caiji Yu Chuli
Contrastive learning， as a self-supervised approach， enables the extraction of target representations from unlabeled SAR images， serving as a critical technique for automatic target recognition （ATR） in SAR.
ZHANG Wenqing, WANG Jing, HUANG Xueqin, TIAN Sirui, HE Cheng, ZHANG Jingdong, LI Hongtao   +6 more
doaj   +1 more source

GFCNet: Contrastive Learning Network with Geography Feature Space Joint Negative Sample Correction for Land Cover Classification

Remote Sensing, 2023
With the continuous improvement in the volume and spatial resolution of remote sensing images, the self-supervised contrastive learning paradigm driven by a large amount of unlabeled data is expected to be a promising solution for large-scale land cover ...
Zhaoyang Zhang, Wenxuan Jing, Haifeng Li, Chao Tao, Yunsheng Zhang   +4 more
doaj   +1 more source

Contrastive-center loss for deep neural networks

, 2017
The deep convolutional neural network(CNN) has significantly raised the performance of image classification and face recognition. Softmax is usually used as supervision, but it only penalizes the classification loss.
Qi, Ce, Su, Fei
core   +1 more source

Graph contrastive learning with implicit augmentations

Neural Networks, 2023
Existing graph contrastive learning methods rely on augmentation techniques based on random perturbations (e.g., randomly adding or dropping edges and nodes). Nevertheless, altering certain edges or nodes can unexpectedly change the graph characteristics, and choosing the optimal perturbing ratio for each dataset requires onerous manual tuning. In this
Huidong Liang, Xingjian Du, Bilei Zhu, Zejun Ma, Ke Chen, Junbin Gao   +5 more
openaire   +3 more sources

Dammarenediol II enhances etoposide‐induced apoptosis by targeting O‐GlcNAc transferase and Akt/GSK3β/mTOR signaling in liver cancer

Molecular Oncology, EarlyView.
Etoposide induces DNA damage, activating p53‐dependent apoptosis via caspase‐3/7, which cleaves PARP1. Dammarenediol II enhances this apoptotic pathway by suppressing O‐GlcNAc transferase activity, further decreasing O‐GlcNAcylation. The reduction in O‐GlcNAc levels boosts p53‐driven apoptosis and influences the Akt/GSK3β/mTOR signaling pathway ...
Jaehoon Lee, Byung‐Cheol Han, Gi‐Bang Koo, Jihye Park, Mijin Kwon, Young Bin Park, Jae‐Mun Choi, Seung‐Ho Lee, Sangho Roh   +8 more
wiley   +1 more source