Search

Your search keyword '"Xu, Huiwen"' showing total 693 results
Start Over Author "Xu, Huiwen"
693 results on '"Xu, Huiwen"'

3. Fast and Accurate Transferability Measurement by Evaluating Intra-class Feature Variance

Author

Xu, Huiwen and Kang, U

Subjects
Computer Science - Machine Learning
Abstract

Given a set of pre-trained models, how can we quickly and accurately find the most useful pre-trained model for a downstream task? Transferability measurement is to quantify how transferable is a pre-trained model learned on a source task to a target task. It is used for quickly ranking pre-trained models for a given task and thus becomes a crucial step for transfer learning. Existing methods measure transferability as the discrimination ability of a source model for a target data before transfer learning, which cannot accurately estimate the fine-tuning performance. Some of them restrict the application of transferability measurement in selecting the best supervised pre-trained models that have classifiers. It is important to have a general method for measuring transferability that can be applied in a variety of situations, such as selecting the best self-supervised pre-trained models that do not have classifiers, and selecting the best transferring layer for a target task. In this work, we propose TMI (TRANSFERABILITY MEASUREMENT WITH INTRA-CLASS FEATURE VARIANCE), a fast and accurate algorithm to measure transferability. We view transferability as the generalization of a pre-trained model on a target task by measuring intra-class feature variance. Intra-class variance evaluates the adaptability of the model to a new task, which measures how transferable the model is. Compared to previous studies that estimate how discriminative the models are, intra-class variance is more accurate than those as it does not require an optimal feature extractor and classifier. Extensive experiments on real-world datasets show that TMI outperforms competitors for selecting the top-5 best models, and exhibits consistently better correlation in 13 out of 17 cases.

Published
2023

10. Fast and Accurate Transferability Measurement for Heterogeneous Multivariate Data

Author

Park, Seungcheol, Xu, Huiwen, Kim, Taehun, Hwang, Inhwan, Kim, Kyung-Jun, and Kang, U

Subjects
Computer Science - Machine Learning, Computer Science - Artificial Intelligence, Statistics - Machine Learning
Abstract

Given a set of heterogeneous source datasets with their classifiers, how can we quickly find the most useful source dataset for a specific target task? We address the problem of measuring transferability between source and target datasets, where the source and the target have different feature spaces and distributions. We propose Transmeter, a fast and accurate method to estimate the transferability of two heterogeneous multivariate datasets. We address three challenges in measuring transferability between two heterogeneous multivariate datasets: reducing time, minimizing domain gap, and extracting meaningful homogeneous representations. To overcome the above issues, we utilize a pre-trained source model, an adversarial network, and an encoder-decoder architecture. Extensive experiments on heterogeneous multivariate datasets show that Transmeter gives the most accurate transferability measurement with up to 10.3 times faster performance than its competitor. We also show that selecting the best source data with Transmeter followed by a full transfer leads to the best transfer accuracy and the fastest running time.

Published
2019

16. Cancer disparities in the context of rurality: risk factors and screening across various U.S. rural classification codes

Author

Hirko, Kelly A., Xu, Huiwen, Rogers, Laura Q., Martin, Michelle Y., Roy, Siddhartha, Kelly, Kimberly M., Christy, Shannon M., Ashing, Kimlin Tam, Yi, Jean C., Lewis-Thames, Marquita W., Meade, Cathy D., Lu, Qian, Gwede, Clement K., Nemeth, Julianna, Ceballos, Rachel M., Menon, Usha, Cueva, Katie, Yeary, Karen, Klesges, Lisa M., Baskin, Monica L., Alcaraz, Kassandra I., and Ford, Sabrina

Published
2022
Full Text
View/download PDF

18. Tabletop single-shot extreme ultraviolet Fourier transform holography of an extended object

Author

Malm, Erik B., Monserud, Nils C., Brown, Christopher G., Wachulak, Przemyslaw W., Xu, Huiwen, Balakrishnan, Ganesh, Chao, Weilun, Anderson, Erik, and Marconi, Mario C.

Subjects
Physics - Optics, Physics - Plasma Physics
Abstract

We demonstrate single and multi-shot Fourier transform holography with the use of a tabletop extreme ultraviolet laser. The reference wave was produced by a Fresnel zone plate with a central opening that allowed the incident beam to illuminate the sample directly. The high reference wave intensity allows for larger objects to be imaged compared to mask-based lensless Fourier transform holography techniques. We obtain a spatial resolution of 169 nm from a single laser pulse and a resolution of 128 nm from an accumulation of 20 laser pulses for an object ~11x11{\mu}m 2 in size. This experiment utilized a tabletop extreme ultraviolet laser that produces a highly coherent ~1.2 ns laser pulse at 46.9 nm wavelength.

Published
2018
Full Text
View/download PDF

24. No evidence of brown adipose tissue activation after 24 weeks of supervised exercise training in young sedentary adults in the ACTIBATE randomized controlled trial

Author

Martinez-Tellez, Borja, Sanchez-Delgado, Guillermo, Acosta, Francisco M., Alcantara, Juan M. A., Amaro-Gahete, Francisco J., Martinez-Avila, Wendy D., Merchan-Ramirez, Elisa, Muñoz-Hernandez, Victoria, Osuna-Prieto, Francisco J., Jurado-Fasoli, Lucas, Xu, Huiwen, Ortiz-Alvarez, Lourdes, Arias-Tellez, María J., Mendez-Gutierrez, Andrea, Labayen, Idoia, Ortega, Francisco B., Schönke, Milena, Rensen, Patrick C. N., Aguilera, Concepción M., Llamas-Elvira, José M., Gil, Ángel, and Ruiz, Jonatan R.

Published
2022
Full Text
View/download PDF

30. Changing landscape of nursing homes serving residents with dementia and mental illnesses

Author

Xu, Huiwen, Intrator, Orna, Culakova, Eva, and Bowblis, John R.

Subjects
Dementia -- Care and treatment, Machine learning -- Usage, Medical care -- Quality management, Nursing homes -- Services -- Evaluation, Mental illness -- Care and treatment, Business, Health care industry
Abstract

Objective: Nursing homes (NHs) are serving an increasing proportion of residents with cognitive issues (e.g., dementia) and mental health conditions. This study aims to: (1) implement unsupervised machine learning to cluster NHs based on residents' dementia and mental health conditions; (2) examine NH staffing related to the clusters; and (3) investigate the association of staffing and NH quality (measured by the number of deficiencies and deficiency scores) in each cluster. Data sources: 2009-2017 Certification and Survey Provider Enhanced Reporting (CASPER) were merged with LTCFocUS.org data on NHs in the United States. Study design: Unsupervised machine learning algorithm (K-means) clustered NHs based on percent residents with dementia, depression, and serious mental illness (SMI, e.g., schizophrenia, anxiety). Panel fixed-effects regressions on deficiency outcomes with staffing-cluster interactions were conducted to examine the effects of staffing on deficiency outcomes in each cluster. Data extraction methods: We identified 110,463 NH-year observations from 14,671 unique NHs using CASPER data. Principal findings: Three clusters were identified: low dementia and mental illnesses (Postacute Cluster); high dementia and depression, but low SMI (Long-stay Cluster); and high dementia and mental illnesses (Cognitive-mental Cluster). From 2009 to 2017, the number of Postacute Cluster NHs increased from 3074 to 5719, while the number of Long-stay Cluster NHs decreased from 6745 to 3058. NHs in Long-stay/Cognitive-mental Clusters reported slightly lower nursing staff hours in 2017. Regressions suggested the effect of increasing staffing on reducing deficiencies is statistically similar across NH clusters. For example, 1 hour increase in registered nurse hours per resident day was associated with -0.67 (standard error [SE] = 0.11), -0.88 (SE = 0.12), and -0.97 (SE = 0.15) deficiencies in Postacute Cluster, Long-stay Cluster, and Cognitive-mental Cluster, respectively. Conclusions: Unsupervised machine learning detected a changing landscape of NH serving residents with dementia and mental illnesses, which requires assuring staffing levels and trainings are suited to residents' needs. KEYWORDS deficiency score, dementia, mental illnesses, nursing homes, staffing, unsupervised machine learning What is known on this topic * Nursing homes are serving very high proportions of residents with dementia and mental illnesses. * Better staffing in nursing homes is generally associated with better quality of care, including a lower number of deficiencies. What this study adds * Over the last decade, significant changes in resident conditions were reflected by a different clustering of dementia and mental health conditions in nursing homes. * Nursing homes that served residents with more dementia and mental health issues had similar or lower staffing levels but were more likely to have a dementia special care unit. * The protective effects of staffing on nursing home quality measured by deficiency scores were similar regardless of nursing home clusters., 1 | INTRODUCTION With the accelerated aging population, nursing homes (NHs) are serving an increasing proportion of residents with cognitive issues (e.g., dementia) and mental health conditions. Previous studies reported [...]

Published
2022
Full Text
View/download PDF

33. Cancer-Related Fatigue

Author

Lin, Po-Ju, Belcher, Elizabeth K., Gilmore, Nikesha J., Hardy, Sara J., Xu, Huiwen, Mustian, Karen M., Kimmick, Gretchen G., editor, Shelby, Rebecca A., editor, and Sutton, Linda M., editor

Published
2021
Full Text
View/download PDF

34. Contributions of Anthropogenic Greenhouse Gases and Aerosol Emissions to Changes in Summer Precipitation Over Southern China.

Author

Xu, Huiwen, Chen, Huopo, He, Wenyue, and Chen, Lefeng

Subjects
GREENHOUSE gases, HUMIDITY, OCEAN temperature, SEA level
Abstract

This study investigates the anthropogenic contribution to the summer precipitation changes over southern China and the underlying physical mechanisms. Observations show a wetting trend over southeastern China (SEC) but a drying trend over southwestern China (SWC) in summer of 1961–2014. The dipole pattern can be reasonably reproduced by the anthropogenic forcing simulations of CMIP6 models but with weak trends under the external natural forcing simulations, suggesting the vital human contribution to the observed changes. Particularly, anthropogenic greenhouse gases (GHG) dominate the wetting trend over SEC, while the drying trend over SWC is primarily attributed to anthropogenic aerosol (AA) emissions. Further analysis shows that the GHG concentrations enhance the subtropical high over the western North Pacific (WNP) via the heterogeneous warming of the sea surface temperature, decrease the sea level pressure over eastern China, and increase the atmospheric moisture, facilitating the moisture flux convergence (MFC) and the precipitation over SEC. The GHG‐induced wetting trend is somewhat offset by the inhibited evaporation due to the AA forcing. For SWC, the decreased precipitation is influenced by the anomalous high pressure from India to WNP, which is closely associated with the enhanced Asian AA emission and the interhemispheric asymmetrical distribution of AA emissions. In the upper troposphere, the uneven AA emissions between South and East Asia and Europe weaken the East Asian summer subtropical jet, resisting the western moisture to SWC. Both factors in the low‐and‐high levels suppress the MFC and precipitation over SWC, counteracted by the thermodynamical effects of GHG forcing. Plain Language Summary: Previous studies have documented the emergence of human signal in changes in precipitation over China, but how the anthropogenic forcing affects the precipitation changes across China is still unclear. This study examines the influence of human activities, such as greenhouse gases (GHG) and anthropogenic aerosol (AA) emissions, on summer precipitation changes in southern China. Observations show a wetting trend in southeastern China (SEC) and a drying trend in southwestern China (SWC) from 1961 to 2014. The simulations indicate that GHG contribute to the wetting trend in SEC, while AA emissions are responsible for the drying trend in SWC. GHG enhance the subtropical high over the WNP, increase atmospheric moisture, and promote precipitation in SEC. In SWC, the high pressures influenced by increased Asian AA emissions weaken moisture flow, leading to decreased precipitation. Uneven AA emissions between regions also impact the transport of moisture. Overall, human activities have a significant impact on the observed changes in summer precipitation in southern China. Key Points: The wetting trend in southeastern China is primarily attributed to the greenhouse gases emission through both dynamic and thermodynamic effectsAerosol emissions are responsible for the drying trend in southwestern China and the dynamic processes dominate the decreased precipitationAerosol induced evaporation decreases precipitation in southeastern China and the drying trend in southwestern China is offset by GHG effect [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

35. The relationship of α-hydroxybutyrate dehydrogenase with 1-year outcomes in patients with intracerebral hemorrhage: A retrospective study

Author

Zhang Limin, Rasha Alsamani, Wu Jianwei, Shi Yijun, Wang Dan, Sun Yuehong, Liu Ziwei, Xu Huiwen, Wang Dongzhi, Zhao Xingquan, and Zhang Guojun

Subjects
intracerebral hemorrhage, outcomes, mortality, α-HBDH, cardiovascular events, Neurology. Diseases of the nervous system, RC346-429
Abstract

Background and aimsCardiac enzymes are recognized as a valuable tool for predicting the prognosis of various cardiovascular diseases. The prognostic value of alpha-hydroxybutyrate dehydrogenase (α-HBDH) in patients with intracerebral hemorrhage (ICH) was ambiguous and not evaluated.MethodsTwo hundred and thirteen Chinese patients with ICH participated in the study from December 2018 to December 2019. Laboratory routine tests and cardiac enzymes, including α-HBDH level, were examined and analyzed. All the patients were classified into two groups by the median value of α-HBDH: B1

Published
2022
Full Text
View/download PDF

36. Evaluation of geriatric assessment and management on the toxic effects of cancer treatment (GAP70+): a cluster-randomised study

Author

Mohile, Supriya G, Mohamed, Mostafa R, Xu, Huiwen, Culakova, Eva, Loh, Kah Poh, Magnuson, Allison, Flannery, Marie A, Obrecht, Spencer, Gilmore, Nikesha, Ramsdale, Erika, Dunne, Richard F, Wildes, Tanya, Plumb, Sandy, Patil, Amita, Wells, Megan, Lowenstein, Lisa, Janelsins, Michelle, Mustian, Karen, Hopkins, Judith O, Berenberg, Jeffrey, Anthony, Navin, and Dale, William

Published
2021
Full Text
View/download PDF

40. Caregiving burden of informal caregivers of older adults with advanced cancer: The effects of rurality and education

Author

Xu, Huiwen, Kadambi, Sindhuja, Mohile, Supriya G., Yang, Shuhan, Kehoe, Lee A., Wells, Megan, Culakova, Eva, Kamen, Charles, Obrecht, Spencer, Mohamed, Mostafa, Gilmore, Nikesha J., Magnuson, Allison, Grossman, Valerie Aarne, Hopkins, Judith O., Geer, Jodi, Berenberg, Jeffrey, Mustian, Karen, Cupertino, Anapaula, Mohile, Nimish, and Loh, Kah Poh

Published
2021
Full Text
View/download PDF

42. The potential health effects associated with electronic-cigarette

Author

Tao, Xiaobo, primary, Zhang, Jiale, additional, Meng, Qianyao, additional, Chu, Junfeng, additional, Zhao, Rongrong, additional, Liu, Yiran, additional, Dong, Yang, additional, Xu, Huiwen, additional, Tian, Tian, additional, Cui, Jiahua, additional, Zhang, Lei, additional, and Chu, Minjie, additional

Published
2023
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results