Results 31 to 40 of about 4,472,392 (304)

Revisiting Knowledge Distillation via Label Smoothing Regularization

Computer Vision and Pattern Recognition, 2020
Knowledge Distillation (KD) aims to distill the knowledge of a cumbersome teacher model into a lightweight student model. Its success is generally attributed to the privileged information on similarities among categories provided by the teacher model ...
Li Yuan, Francis E. H. Tay, Guilin Li, Tao Wang, Jiashi Feng   +4 more
semanticscholar   +1 more source

Unveiling New States in 98Zr: Insights from β-decay and γ-γ Angular-Correlation Studies [PDF]

EPJ Web of Conferences
The nuclear structure of the 98Zr nucleus was studied through the β− decay of 98 Yg.s. at the TRIUMF-ISAC facility. The use of the 8π γ-ray spectrometer with its ancillary detectors SCEPTAR and PACES enabled γ-γ and γ-e− coincidence measurements as well ...
Mashtakov K. R., Garrett P. E., Olaizola B., Andreoiu C., Ball G. C., Bender P., Bildstein V., Chester A., Cross D. S., Dawkins H., Demand G. A., Diaz-Varela A., Deng G., Garnsworthy A. B., Hackman G., Hadinia B., Laffoley A. T., Moukaddam M., Park J., Peters E., Radich A. J., Rajabali M., Rand E. T., Rizwan U., Singh B., Starosta K., Svensson C. E., Voss P., Wang Z-M., Wood J. L., Yates S. W.   +30 more
doaj   +1 more source

Making Risk Minimization Tolerant to Label Noise [PDF]

, 2015
In many applications, the training data, from which one needs to learn a classifier, is corrupted with label noise. Many standard algorithms such as SVM perform poorly in presence of label noise.
Ghosh, Aritra, Manwani, Naresh, Sastry, P. S.   +2 more
core   +1 more source

CNN-RNN: A Unified Framework for Multi-label Image Classification [PDF]

Computer Vision and Pattern Recognition, 2016
While deep convolutional neural networks (CNNs) have shown a great success in single-label image classification, it is important to note that real world images generally contain multiple labels, which could correspond to different objects, scenes ...
Jiang Wang, Yi Yang, Junhua Mao, Zhiheng Huang, Chang Huang, W. Xu   +5 more
semanticscholar   +1 more source

Membership Leakage in Label-Only Exposures [PDF]

Conference on Computer and Communications Security, 2020
Machine learning (ML) has been widely adopted in various privacy-critical applications, e.g., face recognition and medical image analysis. However, recent research has shown that ML models are vulnerable to attacks against their training data. Membership
Zheng Li, Yang Zhang
semanticscholar   +1 more source

Relationships among food label use, motivation, and dietary quality. [PDF]

, 2015
Nutrition information on packaged foods supplies information that aids consumers in meeting the recommendations put forth in the US Dietary Guidelines for Americans such as reducing intake of solid fats and added sugars. It is important to understand how
Applegate, Elizabeth A, Beckett, Laurel A, Cassady, Diana L, Ellwood, Kathleen, Gibson, Tanja N, Miller, Lisa M Soederberg, Wilson, Machelle D   +6 more
core   +3 more sources

Point labeling with sliding labels

Computational Geometry, 1999
zbMATH Open Web Interface contents unavailable due to conflicting licenses.
van Kreveld, M.J., Strijk, T.W., Wolff, A.   +2 more
openaire   +2 more sources

Wasting the Future: The Technological Sublime, Communications Technologies, and E-waste

communication +1, 2012
Literally speaking, e-waste is the future of communications. E-waste is the fastest growing waste stream in the world, much of it communications technologies from cell phones to laptops, televisions to peripherals.
Sebine Label
doaj   +1 more source

Transductive Multi-label Zero-shot Learning [PDF]

, 2014
Zero-shot learning has received increasing interest as a means to alleviate the often prohibitive expense of annotating training data for large scale recognition problems.
Fu, Yanwei, Gong, Shaogang, Hospedales, Tim, Xiang, Tao, Yang, Yongxin   +4 more
core   +2 more sources

Many-to-One Boundary Labeling with Backbones [PDF]

, 2013
In this paper we study \emph{many-to-one boundary labeling with backbone leaders}. In this new many-to-one model, a horizontal backbone reaches out of each label into the feature-enclosing rectangle. Feature points that need to be connected to this label
Bekos, Michael A., Cornelsen, Sabine, Fink, Martin, Hong, Seokhee, Kaufmann, Michael, Nöllenburg, Martin, Rutter, Ignaz, Symvonis, Antonios   +7 more
core   +1 more source