1.Introduction The field of exercise science is experiencing a renaissance,with recent research illuminating the molecular,cellular,and systemic effects of physical activity.This is largely due to the now unequivocal ...1.Introduction The field of exercise science is experiencing a renaissance,with recent research illuminating the molecular,cellular,and systemic effects of physical activity.This is largely due to the now unequivocal evidence that a lack of physical activity,not only has direct effects on the prevalence of non-contagious diseases(NCDs)but has profound additive effects of other risk factors for NCD such as obesity and hypertension.1 The articles in this special topic of Journal of Sport and Health Science(JSHS)are dedicated to research on Exercise biochemistry&metabolism.展开更多
Rheumatoid arthritis(RA)patients face significant psychological challenges alongside physical symptoms,necessitating a comprehensive understanding of how psychological vulnerability and adaptation patterns evolve thro...Rheumatoid arthritis(RA)patients face significant psychological challenges alongside physical symptoms,necessitating a comprehensive understanding of how psychological vulnerability and adaptation patterns evolve throughout the disease course.This review examined 95 studies(2000-2025)from PubMed,Web of Science,and CNKI databases including longitudinal cohorts,randomized controlled trials,and mixed-methods research,to characterize the complex interplay between biological,psychological,and social factors affecting RA patients’mental health.Findings revealed three distinct vulnerability trajectories(45%persistently low,30%fluctuating improvement,25%persistently high)and four adaptation stages,with critical intervention periods occurring 3-6 months postdiagnosis and during disease flares.Multiple factors significantly influence psychological outcomes,including gender(females showing 1.8-fold increased risk),age(younger patients experiencing 42%higher vulnerability),pain intensity,inflammatory markers,and neuroendocrine dysregulation(48%showing cortisol rhythm disruption).Early psychological intervention(within 3 months of diagnosis)demonstrated robust benefits,reducing depression incidence by 42%with effects persisting 24-36 months,while different modalities showed complementary advantages:Cognitive behavioral therapy for depression(Cohen’s d=0.68),mindfulness for pain acceptance(38%improvement),and peer support for meaning reconstruction(25.6%increase).These findings underscore the importance of integrating routine psychological assessment into standard RA care,developing stage-appropriate interventions,and advancing research toward personalized biopsychosocial approaches that address the dynamic psychological dimensions of the disease.展开更多
Domain adaptation aims to reduce the distribution gap between the training data(source domain)and the target data.This enables effective predictions even for domains not seen during training.However,most conventional ...Domain adaptation aims to reduce the distribution gap between the training data(source domain)and the target data.This enables effective predictions even for domains not seen during training.However,most conventional domain adaptation methods assume a single source domain,making them less suitable for modern deep learning settings that rely on diverse and large-scale datasets.To address this limitation,recent research has focused on Multi-Source Domain Adaptation(MSDA),which aims to learn effectively from multiple source domains.In this paper,we propose Efficient Domain Transition for Multi-source(EDTM),a novel and efficient framework designed to tackle two major challenges in existing MSDA approaches:(1)integrating knowledge across different source domains and(2)aligning label distributions between source and target domains.EDTM leverages an ensemble-based classifier expert mechanism to enhance the contribution of source domains that are more similar to the target domain.To further stabilize the learning process and improve performance,we incorporate imitation learning into the training of the target model.In addition,Maximum Classifier Discrepancy(MCD)is employed to align class-wise label distributions between the source and target domains.Experiments were conducted using Digits-Five,one of the most representative benchmark datasets for MSDA.The results show that EDTM consistently outperforms existing methods in terms of average classification accuracy.Notably,EDTM achieved significantly higher performance on target domains such as Modified National Institute of Standards and Technolog with blended background images(MNIST-M)and Street View House Numbers(SVHN)datasets,demonstrating enhanced generalization compared to baseline approaches.Furthermore,an ablation study analyzing the contribution of each loss component validated the effectiveness of the framework,highlighting the importance of each module in achieving optimal performance.展开更多
To address the issue of scarce labeled samples and operational condition variations that degrade the accuracy of fault diagnosis models in variable-condition gearbox fault diagnosis,this paper proposes a semi-supervis...To address the issue of scarce labeled samples and operational condition variations that degrade the accuracy of fault diagnosis models in variable-condition gearbox fault diagnosis,this paper proposes a semi-supervised masked contrastive learning and domain adaptation(SSMCL-DA)method for gearbox fault diagnosis under variable conditions.Initially,during the unsupervised pre-training phase,a dual signal augmentation strategy is devised,which simultaneously applies random masking in the time domain and random scaling in the frequency domain to unlabeled samples,thereby constructing more challenging positive sample pairs to guide the encoder in learning intrinsic features robust to condition variations.Subsequently,a ConvNeXt-Transformer hybrid architecture is employed,integrating the superior local detail modeling capacity of ConvNeXt with the robust global perception capability of Transformer to enhance feature extraction in complex scenarios.Thereafter,a contrastive learning model is constructed with the optimization objective of maximizing feature similarity across different masked instances of the same sample,enabling the extraction of consistent features from multiple masked perspectives and reducing reliance on labeled data.In the final supervised fine-tuning phase,a multi-scale attention mechanism is incorporated for feature rectification,and a domain adaptation module combining Local Maximum Mean Discrepancy(LMMD)with adversarial learning is proposed.This module embodies a dual mechanism:LMMD facilitates fine-grained class-conditional alignment,compelling features of identical fault classes to converge across varying conditions,while the domain discriminator utilizes adversarial training to guide the feature extractor toward learning domain-invariant features.Working in concert,they markedly diminish feature distribution discrepancies induced by changes in load,rotational speed,and other factors,thereby boosting the model’s adaptability to cross-condition scenarios.Experimental evaluations on the WT planetary gearbox dataset and the Case Western Reserve University(CWRU)bearing dataset demonstrate that the SSMCL-DA model effectively identifies multiple fault classes in gearboxes,with diagnostic performance substantially surpassing that of conventional methods.Under cross-condition scenarios,the model attains fault diagnosis accuracies of 99.21%for the WT planetary gearbox and 99.86%for the bearings,respectively.Furthermore,the model exhibits stable generalization capability in cross-device settings.展开更多
Existing elevator fault diagnosis algorithms have limited engineering applicability due to variations in working conditions and differences in equipment structures.To address this limitation,this study proposes an uns...Existing elevator fault diagnosis algorithms have limited engineering applicability due to variations in working conditions and differences in equipment structures.To address this limitation,this study proposes an unsupervised subdomain adaptation method based on a time-frequency feature attention mechanism,LMMD-based subdomain alignment,and contrastive local alignment.This enables the application of the diagnosis model across different working conditions and equipment types.First,a novel time-frequency feature attention mechanism assigns weights to vibration signals of varying dimensions.Second,the time series is transformed to obtain a three-channel time-frequency diagram.This diagram is input into the proposed dimension-segmentation cross-channel multihead self-attention framework to extract high-dimensional frequencydomain fault features.These features are concatenated with the time-domain features to obtain a global feature representation.Then,the extracted high-dimensional features are sent to the classification module to obtain the predicted labels for the source and target domains.Finally,after confidence filtering,the true labels from the source domain and the prediction labels from the target domain are fed into a dynamically weighted multilevel feature alignment module to promote proximity between similar fault features across domains while enhancing separation among different fault types.The validity and superiority of the proposed method were demonstrated through simulation experiments conducted on two types of manned escalator systems under multiple working conditions.For the most challenging transfer task,the proposed method achieved higher accuracy on the target domain test set than DANN,ADDA,C-CLCN,TFA-CCN,and TFA-LCN by 26.87%,24.72%,11.44%,28.94%,and 16.85%,respectively.展开更多
This study explores the cultural,social,and academic adaptation experiences of international students in Wenzhou,China.Based on surveys and interviews with 52 students from 20 countries—predominantly Morocco—the res...This study explores the cultural,social,and academic adaptation experiences of international students in Wenzhou,China.Based on surveys and interviews with 52 students from 20 countries—predominantly Morocco—the research investigates key challenges and coping strategies related to local integration.The findings indicate that while Wenzhou offers a generally supportive academic environment—enhanced by AI integration and practical teaching methods—language barriers continue to hinder students’daily life,academic engagement,and social interactions.Limited Mandarin proficiency made it difficult for many students to build friendships with locals and navigate everyday tasks.Cultural adaptation also presented obstacles,particularly in adjusting to local food and social norms.Despite these challenges,students employed various strategies to facilitate integration,such as attending HSK language courses,watching Chinese media,and initiating conversations with local peers.While most participants described the local community as welcoming,perceptions varied based on individual experiences and language ability.The study highlights the importance of enhanced language support and structured cross-cultural exchange initiatives in improving international students’experiences.It contributes to the broader discourse on international student mobility by offering insights from a second-tier Chinese city,emphasizing the role of institutional practices in shaping adaptation outcomes.展开更多
Understanding the ecological adaptation of tree species can not only reveal the evolutionary potential but also benefit biodiversity conservation under global climate change.Quercus is a keystone genus in Northern Hem...Understanding the ecological adaptation of tree species can not only reveal the evolutionary potential but also benefit biodiversity conservation under global climate change.Quercus is a keystone genus in Northern Hemisphere forests,and its wide distribution in diverse ecosystems and long evolutionary history make it an ideal model for studying the genomic basis of ecological adaptations.Here we used a newly sequenced genome of Quercus gilva,an evergreen oak species from East Asia,with 18 published Fagales genomes to determine how Fagaceae genomes have evolved,identify genomic footprints of ecological adaptability in oaks in general,as well as between evergreen and deciduous oaks.We found that oak species exhibited a higher degree of genomic conservation and stability,as indicated by the absence of large-scale chromosomal structural variations or additional whole-genome duplication events.In addition,we identified expansion and tandem repetitions within gene families that contribute to plant physical and chemical defense(e.g.,cuticle biosynthesis and oxidosqualene cyclase genes),which may represent the foundation for the ecological adaptation of oak species.Circadian rhythm and hormone-related genes may regulate the habits of evergreen and deciduous oaks.This study provides a comprehensive perspective on the ecological adaptations of tree species based on phylogenetic,genome evolutionary,and functional genomic analyses.展开更多
This paper examines the complex trajectory of Chinese medicine’s scientification(科学化)during the late Qing and Republican periods(1850–1949),analyzing how traditional medical knowledge adapted to and negotiated wi...This paper examines the complex trajectory of Chinese medicine’s scientification(科学化)during the late Qing and Republican periods(1850–1949),analyzing how traditional medical knowledge adapted to and negotiated with Western scientific paradigms.Through examination of institutional responses,knowledge transfer networks,and evolving research methodologies,this work demonstrates that the development of scientific Chinese medicine represented a sophisticated process of cultural adaptation rather than simple Westernization.The research identifies three distinct phases—early debates and responses,Japanese influence and knowledge transfers,and research methodologies and institutional development.The 1929 controversy over Yu Yunxiu’s(余云岫)proposal to abolish traditional medicine marked a crucial turning point,catalyzing systematic modernization efforts within the traditional medical community.Japanese influence proved particularly significant through the development of scientific Kampo medicine and the establishment of research networks at institutions.Drawing on Pierre Bourdieu’s concept of scientific fields and Bruno Latour’s actor-network theory,the analysis reveals how different actors negotiated the transformation of traditional medical knowledge within changing social and political contexts.The study demonstrates that Japanese approaches to medical modernization,particularly in pharmacognosy research,provided an alternative model to Western biochemical analysis,emphasizing the preservation of traditional compound formulations while adopting modern scientific methods of converting Chinese medicine to modern.This study contributes to our understanding of medical modernization in East Asia by revealing the sophisticated ways in which traditional knowledge systems adapted to modern scientific requirements while maintaining their essential characteristics.展开更多
The paleoenvironmental changes and adaptation strategies of hominins during the Late Pleistocene are crucial for understanding the evolution,dispersal,and behavioral shifts of early modern humans.Despite South China&#...The paleoenvironmental changes and adaptation strategies of hominins during the Late Pleistocene are crucial for understanding the evolution,dispersal,and behavioral shifts of early modern humans.Despite South China's significance as a nexus for hominin dispersal and handaxe technology diffusion,quantitative reconstructions of paleoenvironments linked to archaeological records remain scarce.The Sandinggai site(96.6-13.3 ka BP)in central South China,with its well-preserved stratigraphy and abundant lithic artefacts,is notable for providing valuable insights.In this study,quantitative reconstruction of the vegetation succession and climate change sequences at the site was conducted using palynological and isotopic data.The results indicated a shift from a warm-temperate evergreen and deciduous broadleaf mixed forest to a temperate deciduous broadleaf forest,with the climate transitioning from warm and humid to cooler and drier conditions.During the early phase,an increase in lithic production suggested favorable conditions for hominin survival.In the later phase,decreased lithic production and the replacement of large handaxe tools by smaller flake tools,indicated that hominins adapted to the cooler,drier climate and more open landscapes through lithic miniaturization.These findings highlight the environment-driven adaptation of lithic technology and hominin behavior,thereby shedding light on human survival adaptation strategies.展开更多
With the rapid evolution of artificial intelligence(AI)technologies,the medical industry is undergoing a profound transformation driven by data intelligence.As the foundational element for intelligent diagnosis,precis...With the rapid evolution of artificial intelligence(AI)technologies,the medical industry is undergoing a profound transformation driven by data intelligence.As the foundational element for intelligent diagnosis,precision prevention,and public health governance,medical data is characterized by massive volume,complex structure,diverse sources,high dimensionality,strong privacy,and high timeliness.Traditional data analysis methods are no longer sufficient to meet the comprehensive requirements of data security,intelligent processing,and decision support.Through techniques such as machine learning,deep learning,natural language processing,and multimodal fusion,AI provides robust technical support for medical data cleaning,governance,mining,and application.At the data level,intelligent algorithms enable the standardization,structuring,and interoperability of medical data,promoting information sharing across medical systems.At the model level,AI supports auxiliary diagnosis and precision treatment through image recognition,medical record analysis,and knowledge graph construction.At the system level,intelligent decision-support platforms continuously enhance the efficiency and accuracy of healthcare services.However,the widespread adoption of AI in medicine still faces challenges such as privacy protection,data security,model interpretability,and the lack of unified industry standards.Based on a systematic review of AI’s key supporting technologies in medical data processing and application,this paper focuses on the compliance challenges and adaptation strategies during industry integration and proposes an adaptation framework centered on“technological trustworthiness,data security,and industry collaboration.”The study provides theoretical and practical insights for promoting the standardized and sustainable development of AI in the healthcare industry.展开更多
In recent years,the heterogeneous SAR image classification task of"training on simulated data and testing on measured data"has garnered increasing attention in the field of Synthetic Aperture Radar Automatic...In recent years,the heterogeneous SAR image classification task of"training on simulated data and testing on measured data"has garnered increasing attention in the field of Synthetic Aperture Radar Automatic Target Recognition(SAR-ATR).Although current mainstream domain adaptation methods have made significant breakthroughs in addressing domain shift problems,the escalating model complexity and task complexity have constrained their deployment in real-world applications.To tackle this challenge,this paper proposes a domain adaptation framework based on linear-kernel Maximum Mean Discrepancy(MMD),integrated with a near-zero-cost pseudo-label denoising technique leveraging deep feature clustering.Our method completely eliminates the need for data augmentation and handcrafted feature design,achieving endto-end pseudo-label self-training.Competitive performance is demonstrated across three typical scenarios in the SAMPLE dataset,with the highest accuracy of 98.65%achieved in ScenarioⅢ.The relevant code is available at:https://github.com/TheGreatTreatsby/SAMPLE_MMD.展开更多
This study evaluates the effectiveness of integrating local wisdom with the use of Alat Pemecah,Peredam Ombak,dan Sedimen Traps(APPOSTRAPS)or Breakers,Wave Dampers,and Sediment Traps in empowering coastal communities ...This study evaluates the effectiveness of integrating local wisdom with the use of Alat Pemecah,Peredam Ombak,dan Sedimen Traps(APPOSTRAPS)or Breakers,Wave Dampers,and Sediment Traps in empowering coastal communities in Karawang,Indonesia,as a strategic response to climate change,coastal erosion,and sustainable ecotourism development.The research aims to assess the combined impact of APPOSTRAPS technology and the Jaga Alam Melalui Pemberdayaan Masyarakat Pesisir(JAM PASIR)or Protecting Nature Through Coastal Community Empowerment program in reducing coastal abrasion,restoring mangrove ecosystems,and fostering sustainable livelihoods.A mixed methods approach was applied,combining quantitative analysis of coastline changes using Geographic Information System(GIS)with Landsat and Sentinel-2 imagery(2022–2024),field surveys using differential GPS(±2 m accuracy),and qualitative methods including 150 interviews,18 months of participant observation,and community documentation.Results show a coastline extension of about 400 m(±15 m),increased ecotourism revenue from IDR 11.25 million per month in 2019 to IDR 90 million in 2024,women’s participation rising from 12%to 68%,and livelihood diversification reaching 110%of the target with 98 families involved.APPOSTRAPS,a patented breakwater and sediment trap made from repurposed tires,combined with the JAM PASIR program covering mangrove-based ecotourism,MSMEs for fishermen’s wives,waste management,and the Masyarakat Sadar Lingkungan and Bencana(MASDARLINA)or Environmentally and Disaster Aware Society system,effectively mitigates erosion and supports economic growth.The study concludes that integrating indigenous knowledge and technology strengthens community resilience and provides a replicable model for sustainable coastal adaptation.展开更多
Introduction Tibetan sheep,economically important animals on the Qinghai–Tibet Plateau,have diversified into numerous local breeds with unique characteristics through prolonged environmental adaptation and selective ...Introduction Tibetan sheep,economically important animals on the Qinghai–Tibet Plateau,have diversified into numerous local breeds with unique characteristics through prolonged environmental adaptation and selective breeding.However,most current research focuses on one or two breeds,and lacks a comprehensive representa-tion of the genetic diversity across multiple Tibetan sheep breeds.This study aims to fill this gap by investigating the genetic structure,diversity and high-altitude adaptation of 6 Tibetan sheep breeds using whole-genome rese-quencing data.Results Six Tibetan sheep breeds were investigated in this study,and whole-genome resequencing data were used to investigate their genetic structure and population diversity.The results showed that the 6 Tibetan sheep breeds exhibited distinct separation in the phylogenetic tree;however,the levels of differentiation among the breeds were minimal,with extensive gene flow observed.Population structure analysis broadly categorized the 6 breeds into 3 distinct ecological types:plateau-type,valley-type and Euler-type.Analysis of unique single-nucleotide polymor-phisms(SNPs)and selective sweeps between Argali and Tibetan sheep revealed that Tibetan sheep domestication was associated primarily with sensory and signal transduction,nutrient absorption and metabolism,and growth and reproductive characteristics.Finally,comprehensive analysis of selective sweep and transcriptome data sug-gested that Tibetan sheep breeds inhabiting different altitudes on the Qinghai–Tibet Plateau adapt by enhancing cardiopulmonary function,regulating body fluid balance through renal reabsorption,and modifying nutrient diges-tion and absorption pathways.Conclusion In this study,we investigated the genetic diversity and population structure of 6 Tibetan sheep breeds in Qinghai Province,China.Additionally,we analyzed the domestication traits and investigated the unique adapta-tion mechanisms residing varying altitudes in the plateau region of Tibetan sheep.This study provides valuable insights into the evolutionary processes of Tibetan sheep in extreme environments.These findings will also contribute to the preservation of genetic diversity and offer a foundation for Tibetan sheep diversity preservation and plateau animal environmental adaptation mechanisms.展开更多
This study analyzes the causes and effects of climate change in the upper Niger River basin and the implementation of local adaptation strategies based on EMS(Environmental Management Systems).It aims to strengthen ec...This study analyzes the causes and effects of climate change in the upper Niger River basin and the implementation of local adaptation strategies based on EMS(Environmental Management Systems).It aims to strengthen ecological resilience and sustainable natural resource management practices through training,awareness-raising,and community participation.The work was conducted in three rural communes in the Kissidougou prefecture,located in the Faranah administrative region.Data collection and analysis tools included questionnaires,GPS devices,digital devices,laptops,and Excel and SPSS software.The methodology employed a participatory and multidisciplinary approach combining a literature review,surveys of 163 respondents,semi-structured interviews with 16 key informants,training for 218 technical staff and local elected officials(30%of whom were women),and awareness-raising activities for 1,800 participants in local languages.Five community forests covering 443.58 hectares were integrated into management plans,concerted,under the coordination of the NGO APARFE.The results show an increase in average temperature(+0.8°C since 1960),a decrease in rainfall(-5.3 mm/month),and increased vulnerability of populations dependent on agriculture.The integration of the EMS(Environmental Management System)has led to improvements in environmental governance,community forest management,awareness of sustainable agricultural practices,and the inclusion of women(51%of participants).In short,the EMS is an effective tool for strengthening community resilience and environmental sustainability.展开更多
As climate change triggers unprecedented ecological shifts,it becomes imperative to understand the genetic underpinnings of species’adaptability.Adaptive introgression significantly contributes to organismal adaptati...As climate change triggers unprecedented ecological shifts,it becomes imperative to understand the genetic underpinnings of species’adaptability.Adaptive introgression significantly contributes to organismal adaptation to new environments by introducing genetic variation across species boundaries.However,despite growing recognition of its importance,the extent to which adaptive introgression has shaped the evolutionary history of closely related species remains poorly understood.Here we employed population genetic analyses of high-throughput sequencing data to investigate the interplay between genetic introgression and local adaptation in three species of spruce trees in the genus Picea(P.asperata,P.crassifolia,and P.meyeri).We find distinct genetic differentiation among these species,despite a substantial gene flow.Crucially,we find bidirectional adaptive introgression between allopatrically distributed species pairs and unearthed dozens of genes linked to stress resilience and flowering time.These candidate genes most likely have promoted adaptability of these spruces to historical environmental changes and may enhance their survival and resilience to future climate changes.Our findings highlight that adaptive introgression could be prevalent and bidirectional in a topographically complex area,and this could have contributed to rich genetic variation and diverse habitat usage by tree species.展开更多
Local adaptation is critical for plant survivals and reproductions in the context of global environmental change.Heterogeneous environments impose various selection pressures that influence the fitness of organisms an...Local adaptation is critical for plant survivals and reproductions in the context of global environmental change.Heterogeneous environments impose various selection pressures that influence the fitness of organisms and leave genomic signatures during the process of adaptation to local environments.However,unveiling the genomic signatures of adaptation still poses a major challenge especially for perennials due to limited genomic resources.Here,we utilized Actinidia eriantha,a Chinese endemic liana,as a model case to detect drivers of local adaptation and adaptive signals through landscape genomics for 311 individuals collected from 25 populations.Our results demonstrated precipitation and solar radiation were two crucial factors influencing the patterns of genetic variations and driving adaptive processes.We further uncovered a set of genes involved in adaptation to heterogeneous environments.Among them,AeERF110 showed high genetic differentiation between populations and was confirmed to be involved in local adaptation via changes in allele frequency along with precipitation(Prec_03)and solar radiation(Srad_03)in native habitats separately,implying that adaptive loci frequently exhibited environmental and geographic signals.In addition,we assessed genetic offsets of populations under four future climate models and revealed that populations from middle and east clusters faced higher risks in adapting to future environments,which should address more attentions.Taken together,our study opens new perspectives for understanding the genetic underpinnings of local adaptation in plants to environmental changes in a more comprehensive fashion and offered the guides on applications in conservation efforts.展开更多
The environment has an important impact on maize(Zea mays L.)production,making it necessary to identify plant adaptation regions that are suitable for different maize varieties.Traditional methods using field trials a...The environment has an important impact on maize(Zea mays L.)production,making it necessary to identify plant adaptation regions that are suitable for different maize varieties.Traditional methods using field trials are costly and restricted to a limited number of areas.Identifying adaptation regions based on climate data has great potential,but a basic understanding and a prediction approach for diverse maize varieties are lacking.Here,we collected a representative dataset comprising 32,840 data points from the National Maize Variety Trial Data Management Platform.We employed three traits to characterize the adaptability of different maize varieties:PH(plant height),DTS(days to silking),and yield.First,we quantified the contributions of variety(V),environment(E),and V×E to variance in the three adaptationrelated traits.The mean contributions of E to variance in PH,DTS,and yield were 54.50%,82.87%,and 75.92%,respectively,suggesting that environmental effects are crucial for phenotype construction.Second,we analyzed correlations between the three traits and three environmental indices:GDD(growing degree days),PRE(precipitation),and SSD(sunshine duration).The highest absolute correlation coefficients between phenotypes and environmental indices were 0.15–0.69 at the whole-data level.To predict variety adaptation on a national scale,we modeled the three traits using environmental indices and best linear unbiased predictors(BLUPs)via the random forest algorithm.The predictive abilities of our models for PH,DTS,and yield were 0.90(MAE=9.95 cm),0.99(MAE=1.09 d),and 0.95(MAE=0.55 t ha^(−1)),respectively,indicating that our proposed framework can predict adaptationrelated traits for diverse maize varieties in China.展开更多
Conventional model transfer techniques,requiring the labelled source data,are not applicable in the privacy-protected medical fields.For the challenging scenarios,recent source data-free domain adaptation(SFDA)has bec...Conventional model transfer techniques,requiring the labelled source data,are not applicable in the privacy-protected medical fields.For the challenging scenarios,recent source data-free domain adaptation(SFDA)has become a mainstream solution but losing focus on the inter-sample class information.This paper proposes a new Credible Local Context Representation approach for SFDA.Our main idea is to exploit the credible local context for more discriminative representation.Specifically,we enhance the source model's discrimination by information regulating.To capture the context,a discovery method is developed that performs fixed steps walking in deep space and takes the credible features in this path as the context.In the epoch-wise adaptation,deep clustering-like training is conducted with two major updates.First,the context for all target data is constructed and then the context-fused pseudo-labels providing semantic guidance are generated.Second,for each target data,a weighting fusion on its context forms the anchored neighbourhood structure;thus,the deep clustering is switched from individual-based to coarse-grained.Also,a new regularisation building is developed on the anchored neighbourhood to drive the deep coarse-grained learning.Experiments on three benchmarks indicate that the proposed method can achieve stateof-the-art results.展开更多
文摘1.Introduction The field of exercise science is experiencing a renaissance,with recent research illuminating the molecular,cellular,and systemic effects of physical activity.This is largely due to the now unequivocal evidence that a lack of physical activity,not only has direct effects on the prevalence of non-contagious diseases(NCDs)but has profound additive effects of other risk factors for NCD such as obesity and hypertension.1 The articles in this special topic of Journal of Sport and Health Science(JSHS)are dedicated to research on Exercise biochemistry&metabolism.
基金Supported by Chongqing Health Commission and Chongqing Science and Technology Bureau,No.2023MSXM182。
文摘Rheumatoid arthritis(RA)patients face significant psychological challenges alongside physical symptoms,necessitating a comprehensive understanding of how psychological vulnerability and adaptation patterns evolve throughout the disease course.This review examined 95 studies(2000-2025)from PubMed,Web of Science,and CNKI databases including longitudinal cohorts,randomized controlled trials,and mixed-methods research,to characterize the complex interplay between biological,psychological,and social factors affecting RA patients’mental health.Findings revealed three distinct vulnerability trajectories(45%persistently low,30%fluctuating improvement,25%persistently high)and four adaptation stages,with critical intervention periods occurring 3-6 months postdiagnosis and during disease flares.Multiple factors significantly influence psychological outcomes,including gender(females showing 1.8-fold increased risk),age(younger patients experiencing 42%higher vulnerability),pain intensity,inflammatory markers,and neuroendocrine dysregulation(48%showing cortisol rhythm disruption).Early psychological intervention(within 3 months of diagnosis)demonstrated robust benefits,reducing depression incidence by 42%with effects persisting 24-36 months,while different modalities showed complementary advantages:Cognitive behavioral therapy for depression(Cohen’s d=0.68),mindfulness for pain acceptance(38%improvement),and peer support for meaning reconstruction(25.6%increase).These findings underscore the importance of integrating routine psychological assessment into standard RA care,developing stage-appropriate interventions,and advancing research toward personalized biopsychosocial approaches that address the dynamic psychological dimensions of the disease.
基金supported by the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(No.RS-2024-00406320)the Institute of Information&Communica-tions Technology Planning&Evaluation(IITP)-Innovative Human Resource Development for Local Intellectualization Program Grant funded by the Korea government(MSIT)(IITP-2026-RS-2023-00259678).
文摘Domain adaptation aims to reduce the distribution gap between the training data(source domain)and the target data.This enables effective predictions even for domains not seen during training.However,most conventional domain adaptation methods assume a single source domain,making them less suitable for modern deep learning settings that rely on diverse and large-scale datasets.To address this limitation,recent research has focused on Multi-Source Domain Adaptation(MSDA),which aims to learn effectively from multiple source domains.In this paper,we propose Efficient Domain Transition for Multi-source(EDTM),a novel and efficient framework designed to tackle two major challenges in existing MSDA approaches:(1)integrating knowledge across different source domains and(2)aligning label distributions between source and target domains.EDTM leverages an ensemble-based classifier expert mechanism to enhance the contribution of source domains that are more similar to the target domain.To further stabilize the learning process and improve performance,we incorporate imitation learning into the training of the target model.In addition,Maximum Classifier Discrepancy(MCD)is employed to align class-wise label distributions between the source and target domains.Experiments were conducted using Digits-Five,one of the most representative benchmark datasets for MSDA.The results show that EDTM consistently outperforms existing methods in terms of average classification accuracy.Notably,EDTM achieved significantly higher performance on target domains such as Modified National Institute of Standards and Technolog with blended background images(MNIST-M)and Street View House Numbers(SVHN)datasets,demonstrating enhanced generalization compared to baseline approaches.Furthermore,an ablation study analyzing the contribution of each loss component validated the effectiveness of the framework,highlighting the importance of each module in achieving optimal performance.
基金supported by the National Natural Science Foundation of China Funded Project(Project Name:Research on Robust Adaptive Allocation Mechanism of Human Machine Co-Driving System Based on NMS Features,Project Approval Number:52172381).
文摘To address the issue of scarce labeled samples and operational condition variations that degrade the accuracy of fault diagnosis models in variable-condition gearbox fault diagnosis,this paper proposes a semi-supervised masked contrastive learning and domain adaptation(SSMCL-DA)method for gearbox fault diagnosis under variable conditions.Initially,during the unsupervised pre-training phase,a dual signal augmentation strategy is devised,which simultaneously applies random masking in the time domain and random scaling in the frequency domain to unlabeled samples,thereby constructing more challenging positive sample pairs to guide the encoder in learning intrinsic features robust to condition variations.Subsequently,a ConvNeXt-Transformer hybrid architecture is employed,integrating the superior local detail modeling capacity of ConvNeXt with the robust global perception capability of Transformer to enhance feature extraction in complex scenarios.Thereafter,a contrastive learning model is constructed with the optimization objective of maximizing feature similarity across different masked instances of the same sample,enabling the extraction of consistent features from multiple masked perspectives and reducing reliance on labeled data.In the final supervised fine-tuning phase,a multi-scale attention mechanism is incorporated for feature rectification,and a domain adaptation module combining Local Maximum Mean Discrepancy(LMMD)with adversarial learning is proposed.This module embodies a dual mechanism:LMMD facilitates fine-grained class-conditional alignment,compelling features of identical fault classes to converge across varying conditions,while the domain discriminator utilizes adversarial training to guide the feature extractor toward learning domain-invariant features.Working in concert,they markedly diminish feature distribution discrepancies induced by changes in load,rotational speed,and other factors,thereby boosting the model’s adaptability to cross-condition scenarios.Experimental evaluations on the WT planetary gearbox dataset and the Case Western Reserve University(CWRU)bearing dataset demonstrate that the SSMCL-DA model effectively identifies multiple fault classes in gearboxes,with diagnostic performance substantially surpassing that of conventional methods.Under cross-condition scenarios,the model attains fault diagnosis accuracies of 99.21%for the WT planetary gearbox and 99.86%for the bearings,respectively.Furthermore,the model exhibits stable generalization capability in cross-device settings.
基金supported by the National Natural Science Foundation of China(Grant Nos.52375255,51935007)the Shanghai Rising-Star Program(Grant No.24QB2705000)。
文摘Existing elevator fault diagnosis algorithms have limited engineering applicability due to variations in working conditions and differences in equipment structures.To address this limitation,this study proposes an unsupervised subdomain adaptation method based on a time-frequency feature attention mechanism,LMMD-based subdomain alignment,and contrastive local alignment.This enables the application of the diagnosis model across different working conditions and equipment types.First,a novel time-frequency feature attention mechanism assigns weights to vibration signals of varying dimensions.Second,the time series is transformed to obtain a three-channel time-frequency diagram.This diagram is input into the proposed dimension-segmentation cross-channel multihead self-attention framework to extract high-dimensional frequencydomain fault features.These features are concatenated with the time-domain features to obtain a global feature representation.Then,the extracted high-dimensional features are sent to the classification module to obtain the predicted labels for the source and target domains.Finally,after confidence filtering,the true labels from the source domain and the prediction labels from the target domain are fed into a dynamically weighted multilevel feature alignment module to promote proximity between similar fault features across domains while enhancing separation among different fault types.The validity and superiority of the proposed method were demonstrated through simulation experiments conducted on two types of manned escalator systems under multiple working conditions.For the most challenging transfer task,the proposed method achieved higher accuracy on the target domain test set than DANN,ADDA,C-CLCN,TFA-CCN,and TFA-LCN by 26.87%,24.72%,11.44%,28.94%,and 16.85%,respectively.
基金supported by Cultural and Ideological Progress Director Center of Ouhai District of Wenzhou(2024-135F).
文摘This study explores the cultural,social,and academic adaptation experiences of international students in Wenzhou,China.Based on surveys and interviews with 52 students from 20 countries—predominantly Morocco—the research investigates key challenges and coping strategies related to local integration.The findings indicate that while Wenzhou offers a generally supportive academic environment—enhanced by AI integration and practical teaching methods—language barriers continue to hinder students’daily life,academic engagement,and social interactions.Limited Mandarin proficiency made it difficult for many students to build friendships with locals and navigate everyday tasks.Cultural adaptation also presented obstacles,particularly in adjusting to local food and social norms.Despite these challenges,students employed various strategies to facilitate integration,such as attending HSK language courses,watching Chinese media,and initiating conversations with local peers.While most participants described the local community as welcoming,perceptions varied based on individual experiences and language ability.The study highlights the importance of enhanced language support and structured cross-cultural exchange initiatives in improving international students’experiences.It contributes to the broader discourse on international student mobility by offering insights from a second-tier Chinese city,emphasizing the role of institutional practices in shaping adaptation outcomes.
基金supported by the National Natural Science Foundation of China(No.31901217)the Special Fund for Scientific Research of Shanghai Landscaping and City Appearance Administrative Bureau(grant numbers G192422,G242414,and G242416).
文摘Understanding the ecological adaptation of tree species can not only reveal the evolutionary potential but also benefit biodiversity conservation under global climate change.Quercus is a keystone genus in Northern Hemisphere forests,and its wide distribution in diverse ecosystems and long evolutionary history make it an ideal model for studying the genomic basis of ecological adaptations.Here we used a newly sequenced genome of Quercus gilva,an evergreen oak species from East Asia,with 18 published Fagales genomes to determine how Fagaceae genomes have evolved,identify genomic footprints of ecological adaptability in oaks in general,as well as between evergreen and deciduous oaks.We found that oak species exhibited a higher degree of genomic conservation and stability,as indicated by the absence of large-scale chromosomal structural variations or additional whole-genome duplication events.In addition,we identified expansion and tandem repetitions within gene families that contribute to plant physical and chemical defense(e.g.,cuticle biosynthesis and oxidosqualene cyclase genes),which may represent the foundation for the ecological adaptation of oak species.Circadian rhythm and hormone-related genes may regulate the habits of evergreen and deciduous oaks.This study provides a comprehensive perspective on the ecological adaptations of tree species based on phylogenetic,genome evolutionary,and functional genomic analyses.
文摘This paper examines the complex trajectory of Chinese medicine’s scientification(科学化)during the late Qing and Republican periods(1850–1949),analyzing how traditional medical knowledge adapted to and negotiated with Western scientific paradigms.Through examination of institutional responses,knowledge transfer networks,and evolving research methodologies,this work demonstrates that the development of scientific Chinese medicine represented a sophisticated process of cultural adaptation rather than simple Westernization.The research identifies three distinct phases—early debates and responses,Japanese influence and knowledge transfers,and research methodologies and institutional development.The 1929 controversy over Yu Yunxiu’s(余云岫)proposal to abolish traditional medicine marked a crucial turning point,catalyzing systematic modernization efforts within the traditional medical community.Japanese influence proved particularly significant through the development of scientific Kampo medicine and the establishment of research networks at institutions.Drawing on Pierre Bourdieu’s concept of scientific fields and Bruno Latour’s actor-network theory,the analysis reveals how different actors negotiated the transformation of traditional medical knowledge within changing social and political contexts.The study demonstrates that Japanese approaches to medical modernization,particularly in pharmacognosy research,provided an alternative model to Western biochemical analysis,emphasizing the preservation of traditional compound formulations while adopting modern scientific methods of converting Chinese medicine to modern.This study contributes to our understanding of medical modernization in East Asia by revealing the sophisticated ways in which traditional knowledge systems adapted to modern scientific requirements while maintaining their essential characteristics.
基金National Natural Science Foundation of China,No.42471185,No.T2192952National Key Research and Development Program of China,No.2022YFF0801502。
文摘The paleoenvironmental changes and adaptation strategies of hominins during the Late Pleistocene are crucial for understanding the evolution,dispersal,and behavioral shifts of early modern humans.Despite South China's significance as a nexus for hominin dispersal and handaxe technology diffusion,quantitative reconstructions of paleoenvironments linked to archaeological records remain scarce.The Sandinggai site(96.6-13.3 ka BP)in central South China,with its well-preserved stratigraphy and abundant lithic artefacts,is notable for providing valuable insights.In this study,quantitative reconstruction of the vegetation succession and climate change sequences at the site was conducted using palynological and isotopic data.The results indicated a shift from a warm-temperate evergreen and deciduous broadleaf mixed forest to a temperate deciduous broadleaf forest,with the climate transitioning from warm and humid to cooler and drier conditions.During the early phase,an increase in lithic production suggested favorable conditions for hominin survival.In the later phase,decreased lithic production and the replacement of large handaxe tools by smaller flake tools,indicated that hominins adapted to the cooler,drier climate and more open landscapes through lithic miniaturization.These findings highlight the environment-driven adaptation of lithic technology and hominin behavior,thereby shedding light on human survival adaptation strategies.
文摘With the rapid evolution of artificial intelligence(AI)technologies,the medical industry is undergoing a profound transformation driven by data intelligence.As the foundational element for intelligent diagnosis,precision prevention,and public health governance,medical data is characterized by massive volume,complex structure,diverse sources,high dimensionality,strong privacy,and high timeliness.Traditional data analysis methods are no longer sufficient to meet the comprehensive requirements of data security,intelligent processing,and decision support.Through techniques such as machine learning,deep learning,natural language processing,and multimodal fusion,AI provides robust technical support for medical data cleaning,governance,mining,and application.At the data level,intelligent algorithms enable the standardization,structuring,and interoperability of medical data,promoting information sharing across medical systems.At the model level,AI supports auxiliary diagnosis and precision treatment through image recognition,medical record analysis,and knowledge graph construction.At the system level,intelligent decision-support platforms continuously enhance the efficiency and accuracy of healthcare services.However,the widespread adoption of AI in medicine still faces challenges such as privacy protection,data security,model interpretability,and the lack of unified industry standards.Based on a systematic review of AI’s key supporting technologies in medical data processing and application,this paper focuses on the compliance challenges and adaptation strategies during industry integration and proposes an adaptation framework centered on“technological trustworthiness,data security,and industry collaboration.”The study provides theoretical and practical insights for promoting the standardized and sustainable development of AI in the healthcare industry.
文摘In recent years,the heterogeneous SAR image classification task of"training on simulated data and testing on measured data"has garnered increasing attention in the field of Synthetic Aperture Radar Automatic Target Recognition(SAR-ATR).Although current mainstream domain adaptation methods have made significant breakthroughs in addressing domain shift problems,the escalating model complexity and task complexity have constrained their deployment in real-world applications.To tackle this challenge,this paper proposes a domain adaptation framework based on linear-kernel Maximum Mean Discrepancy(MMD),integrated with a near-zero-cost pseudo-label denoising technique leveraging deep feature clustering.Our method completely eliminates the need for data augmentation and handcrafted feature design,achieving endto-end pseudo-label self-training.Competitive performance is demonstrated across three typical scenarios in the SAMPLE dataset,with the highest accuracy of 98.65%achieved in ScenarioⅢ.The relevant code is available at:https://github.com/TheGreatTreatsby/SAMPLE_MMD.
文摘This study evaluates the effectiveness of integrating local wisdom with the use of Alat Pemecah,Peredam Ombak,dan Sedimen Traps(APPOSTRAPS)or Breakers,Wave Dampers,and Sediment Traps in empowering coastal communities in Karawang,Indonesia,as a strategic response to climate change,coastal erosion,and sustainable ecotourism development.The research aims to assess the combined impact of APPOSTRAPS technology and the Jaga Alam Melalui Pemberdayaan Masyarakat Pesisir(JAM PASIR)or Protecting Nature Through Coastal Community Empowerment program in reducing coastal abrasion,restoring mangrove ecosystems,and fostering sustainable livelihoods.A mixed methods approach was applied,combining quantitative analysis of coastline changes using Geographic Information System(GIS)with Landsat and Sentinel-2 imagery(2022–2024),field surveys using differential GPS(±2 m accuracy),and qualitative methods including 150 interviews,18 months of participant observation,and community documentation.Results show a coastline extension of about 400 m(±15 m),increased ecotourism revenue from IDR 11.25 million per month in 2019 to IDR 90 million in 2024,women’s participation rising from 12%to 68%,and livelihood diversification reaching 110%of the target with 98 families involved.APPOSTRAPS,a patented breakwater and sediment trap made from repurposed tires,combined with the JAM PASIR program covering mangrove-based ecotourism,MSMEs for fishermen’s wives,waste management,and the Masyarakat Sadar Lingkungan and Bencana(MASDARLINA)or Environmentally and Disaster Aware Society system,effectively mitigates erosion and supports economic growth.The study concludes that integrating indigenous knowledge and technology strengthens community resilience and provides a replicable model for sustainable coastal adaptation.
基金supported by the Natural Science Foundation of Qinghai Province(No.2022-ZJ-901)the National Breeding Joint Research Project。
文摘Introduction Tibetan sheep,economically important animals on the Qinghai–Tibet Plateau,have diversified into numerous local breeds with unique characteristics through prolonged environmental adaptation and selective breeding.However,most current research focuses on one or two breeds,and lacks a comprehensive representa-tion of the genetic diversity across multiple Tibetan sheep breeds.This study aims to fill this gap by investigating the genetic structure,diversity and high-altitude adaptation of 6 Tibetan sheep breeds using whole-genome rese-quencing data.Results Six Tibetan sheep breeds were investigated in this study,and whole-genome resequencing data were used to investigate their genetic structure and population diversity.The results showed that the 6 Tibetan sheep breeds exhibited distinct separation in the phylogenetic tree;however,the levels of differentiation among the breeds were minimal,with extensive gene flow observed.Population structure analysis broadly categorized the 6 breeds into 3 distinct ecological types:plateau-type,valley-type and Euler-type.Analysis of unique single-nucleotide polymor-phisms(SNPs)and selective sweeps between Argali and Tibetan sheep revealed that Tibetan sheep domestication was associated primarily with sensory and signal transduction,nutrient absorption and metabolism,and growth and reproductive characteristics.Finally,comprehensive analysis of selective sweep and transcriptome data sug-gested that Tibetan sheep breeds inhabiting different altitudes on the Qinghai–Tibet Plateau adapt by enhancing cardiopulmonary function,regulating body fluid balance through renal reabsorption,and modifying nutrient diges-tion and absorption pathways.Conclusion In this study,we investigated the genetic diversity and population structure of 6 Tibetan sheep breeds in Qinghai Province,China.Additionally,we analyzed the domestication traits and investigated the unique adapta-tion mechanisms residing varying altitudes in the plateau region of Tibetan sheep.This study provides valuable insights into the evolutionary processes of Tibetan sheep in extreme environments.These findings will also contribute to the preservation of genetic diversity and offer a foundation for Tibetan sheep diversity preservation and plateau animal environmental adaptation mechanisms.
文摘This study analyzes the causes and effects of climate change in the upper Niger River basin and the implementation of local adaptation strategies based on EMS(Environmental Management Systems).It aims to strengthen ecological resilience and sustainable natural resource management practices through training,awareness-raising,and community participation.The work was conducted in three rural communes in the Kissidougou prefecture,located in the Faranah administrative region.Data collection and analysis tools included questionnaires,GPS devices,digital devices,laptops,and Excel and SPSS software.The methodology employed a participatory and multidisciplinary approach combining a literature review,surveys of 163 respondents,semi-structured interviews with 16 key informants,training for 218 technical staff and local elected officials(30%of whom were women),and awareness-raising activities for 1,800 participants in local languages.Five community forests covering 443.58 hectares were integrated into management plans,concerted,under the coordination of the NGO APARFE.The results show an increase in average temperature(+0.8°C since 1960),a decrease in rainfall(-5.3 mm/month),and increased vulnerability of populations dependent on agriculture.The integration of the EMS(Environmental Management System)has led to improvements in environmental governance,community forest management,awareness of sustainable agricultural practices,and the inclusion of women(51%of participants).In short,the EMS is an effective tool for strengthening community resilience and environmental sustainability.
基金the Project of Qinghai provincial central government guides local funds for science and technology development(2024ZY005).
文摘As climate change triggers unprecedented ecological shifts,it becomes imperative to understand the genetic underpinnings of species’adaptability.Adaptive introgression significantly contributes to organismal adaptation to new environments by introducing genetic variation across species boundaries.However,despite growing recognition of its importance,the extent to which adaptive introgression has shaped the evolutionary history of closely related species remains poorly understood.Here we employed population genetic analyses of high-throughput sequencing data to investigate the interplay between genetic introgression and local adaptation in three species of spruce trees in the genus Picea(P.asperata,P.crassifolia,and P.meyeri).We find distinct genetic differentiation among these species,despite a substantial gene flow.Crucially,we find bidirectional adaptive introgression between allopatrically distributed species pairs and unearthed dozens of genes linked to stress resilience and flowering time.These candidate genes most likely have promoted adaptability of these spruces to historical environmental changes and may enhance their survival and resilience to future climate changes.Our findings highlight that adaptive introgression could be prevalent and bidirectional in a topographically complex area,and this could have contributed to rich genetic variation and diverse habitat usage by tree species.
基金funded by the National Natural Science Foundation of China(grants number 32070377 and 31770374)Science Fund for Creative Research Groups of the Natural Science Foundation of Hubei Province(2024AFA035).
文摘Local adaptation is critical for plant survivals and reproductions in the context of global environmental change.Heterogeneous environments impose various selection pressures that influence the fitness of organisms and leave genomic signatures during the process of adaptation to local environments.However,unveiling the genomic signatures of adaptation still poses a major challenge especially for perennials due to limited genomic resources.Here,we utilized Actinidia eriantha,a Chinese endemic liana,as a model case to detect drivers of local adaptation and adaptive signals through landscape genomics for 311 individuals collected from 25 populations.Our results demonstrated precipitation and solar radiation were two crucial factors influencing the patterns of genetic variations and driving adaptive processes.We further uncovered a set of genes involved in adaptation to heterogeneous environments.Among them,AeERF110 showed high genetic differentiation between populations and was confirmed to be involved in local adaptation via changes in allele frequency along with precipitation(Prec_03)and solar radiation(Srad_03)in native habitats separately,implying that adaptive loci frequently exhibited environmental and geographic signals.In addition,we assessed genetic offsets of populations under four future climate models and revealed that populations from middle and east clusters faced higher risks in adapting to future environments,which should address more attentions.Taken together,our study opens new perspectives for understanding the genetic underpinnings of local adaptation in plants to environmental changes in a more comprehensive fashion and offered the guides on applications in conservation efforts.
基金funded by the National Science and Technology Major Project(2022ZD0115703)the Beijing Postdoctoral Research Foundation(2023-ZZ-116).
文摘The environment has an important impact on maize(Zea mays L.)production,making it necessary to identify plant adaptation regions that are suitable for different maize varieties.Traditional methods using field trials are costly and restricted to a limited number of areas.Identifying adaptation regions based on climate data has great potential,but a basic understanding and a prediction approach for diverse maize varieties are lacking.Here,we collected a representative dataset comprising 32,840 data points from the National Maize Variety Trial Data Management Platform.We employed three traits to characterize the adaptability of different maize varieties:PH(plant height),DTS(days to silking),and yield.First,we quantified the contributions of variety(V),environment(E),and V×E to variance in the three adaptationrelated traits.The mean contributions of E to variance in PH,DTS,and yield were 54.50%,82.87%,and 75.92%,respectively,suggesting that environmental effects are crucial for phenotype construction.Second,we analyzed correlations between the three traits and three environmental indices:GDD(growing degree days),PRE(precipitation),and SSD(sunshine duration).The highest absolute correlation coefficients between phenotypes and environmental indices were 0.15–0.69 at the whole-data level.To predict variety adaptation on a national scale,we modeled the three traits using environmental indices and best linear unbiased predictors(BLUPs)via the random forest algorithm.The predictive abilities of our models for PH,DTS,and yield were 0.90(MAE=9.95 cm),0.99(MAE=1.09 d),and 0.95(MAE=0.55 t ha^(−1)),respectively,indicating that our proposed framework can predict adaptationrelated traits for diverse maize varieties in China.
基金National Key R&D Program of China,Grant/Award Numbers:2018YFE0203900,2020YFB1313600German Research Foundation,Hamburg Landesforschungsförderungsprojekt Cross,Grant/Award Number:Sonderforschungsbereich Transregio 169+2 种基金Shanghai Artificial Intelligence Innovation Development Special Support Project,Grant/Award Number:3920365001Horizon2020 RISE project STEP2DYNA,Grant/Award Number:691154National Natural Science Foundation of China,Grant/Award Numbers:61773083,62206168,62276048,U1813202。
文摘Conventional model transfer techniques,requiring the labelled source data,are not applicable in the privacy-protected medical fields.For the challenging scenarios,recent source data-free domain adaptation(SFDA)has become a mainstream solution but losing focus on the inter-sample class information.This paper proposes a new Credible Local Context Representation approach for SFDA.Our main idea is to exploit the credible local context for more discriminative representation.Specifically,we enhance the source model's discrimination by information regulating.To capture the context,a discovery method is developed that performs fixed steps walking in deep space and takes the credible features in this path as the context.In the epoch-wise adaptation,deep clustering-like training is conducted with two major updates.First,the context for all target data is constructed and then the context-fused pseudo-labels providing semantic guidance are generated.Second,for each target data,a weighting fusion on its context forms the anchored neighbourhood structure;thus,the deep clustering is switched from individual-based to coarse-grained.Also,a new regularisation building is developed on the anchored neighbourhood to drive the deep coarse-grained learning.Experiments on three benchmarks indicate that the proposed method can achieve stateof-the-art results.
