In the very beginning,the Computer Laboratory of the University of Cambridge was founded to provide computing service for different disciplines across the university.As computer science developed as a discipline in it...In the very beginning,the Computer Laboratory of the University of Cambridge was founded to provide computing service for different disciplines across the university.As computer science developed as a discipline in its own right,boundaries necessarily arose between it and other disciplines,in a way that is now often detrimental to progress.Therefore,it is necessary to reinvigorate the relationship between computer science and other academic disciplines and celebrate exploration and creativity in research.To do this,the structures of the academic department have to act as supporting scaffolding rather than barriers.Some examples are given that show the efforts being made at the University of Cambridge to approach this problem.展开更多
Spark performs excellently in large-scale data-parallel computing and iterative processing.However,with the increase in data size and program complexity,the default scheduling strategy has difficultymeeting the demand...Spark performs excellently in large-scale data-parallel computing and iterative processing.However,with the increase in data size and program complexity,the default scheduling strategy has difficultymeeting the demands of resource utilization and performance optimization.Scheduling strategy optimization,as a key direction for improving Spark’s execution efficiency,has attracted widespread attention.This paper first introduces the basic theories of Spark,compares several default scheduling strategies,and discusses common scheduling performance evaluation indicators and factors affecting scheduling efficiency.Subsequently,existing scheduling optimization schemes are summarized based on three scheduling modes:load characteristics,cluster characteristics,and matching of both,and representative algorithms are analyzed in terms of performance indicators and applicable scenarios,comparing the advantages and disadvantages of different scheduling modes.The article also explores in detail the integration of Spark scheduling strategies with specific application scenarios and the challenges in production environments.Finally,the limitations of the existing schemes are analyzed,and prospects are envisioned.展开更多
BACKGROUND Although thymopoietin(TMPO)has been elucidated to be overexpressed in cancers,its underlying mechanisms are not yet fully understood.AIM To investigate the expression and clinical significance of TMPO in pa...BACKGROUND Although thymopoietin(TMPO)has been elucidated to be overexpressed in cancers,its underlying mechanisms are not yet fully understood.AIM To investigate the expression and clinical significance of TMPO in papillary thyroid carcinoma(PTC).METHODS Databases such as Gene Expression Omnibus,The Cancer Genome Atlas Proand summary receiver operating characteristic curves were plotted to evaluate diagnostic performance.A Gene Set Enrichment Analysis enrichment analysis was conducted to identify TMPO-related signaling pathways.A protein interaction network was constructed to identify hub genes.The impact of TMPO on PTC cell proliferation and the effects of its knockout were analyzed using clustered regularly interspaced short palindromic repeats(CRISPR)knockout screening and the Cancer Cell Line Encyclopedia database.RESULTS The TMPO protein was significantly overexpressed in PTC tissues,primarily localized in the cytoplasm and nuclear membrane.The mRNA level analysis showed mild overexpression of TMPO in PTC tissues,with a certain discriminatory value(area under the curve=0.66).TMPO may promote cancer through involvement in cell adhesion,focal adhesion,leukocyte migration,and multiple cancer-related signaling pathways.Additionally,CRISPR gene knockout experiments confirmed that TMPO knockout significantly inhibited the proliferation of PTC cell lines,indicating its important role in tumor growth.CONCLUSION TMPO is overexpressed in PTC and may serve as a therapeutic target and molecular biomarker for PTC.展开更多
At the panel session of the 3rd Global Forum on the Development of Computer Science,attendees had an opportunity to deliberate recent issues affecting computer science departments as a result of the recent growth in t...At the panel session of the 3rd Global Forum on the Development of Computer Science,attendees had an opportunity to deliberate recent issues affecting computer science departments as a result of the recent growth in the field.6 heads of university computer science departments participated in the discussions,including the moderator,Professor Andrew Yao.The first issue was how universities are managing the growing number of applicants in addition to swelling class sizes.Several approaches were suggested,including increasing faculty hiring,implementing scalable teaching tools,and working closer with other departments through degree programs that integrate computer science with other fields.The second issue was about the position and role of computer science within broader science.Participants generally agreed that all fields are increasingly relying on computer science techniques,and that effectively disseminating these techniques to others is a key to unlocking broader scientific progress.展开更多
Extracting valuable information frombiomedical texts is one of the current research hotspots of concern to a wide range of scholars.The biomedical corpus contains numerous complex long sentences and overlapping relati...Extracting valuable information frombiomedical texts is one of the current research hotspots of concern to a wide range of scholars.The biomedical corpus contains numerous complex long sentences and overlapping relational triples,making most generalized domain joint modeling methods difficult to apply effectively in this field.For a complex semantic environment in biomedical texts,in this paper,we propose a novel perspective to perform joint entity and relation extraction;existing studies divide the relation triples into several steps or modules.However,the three elements in the relation triples are interdependent and inseparable,so we regard joint extraction as a tripartite classification problem.At the same time,fromthe perspective of triple classification,we design amulti-granularity 2D convolution to refine the word pair table and better utilize the dependencies between biomedical word pairs.Finally,we use a biaffine predictor to assist in predicting the labels of word pairs for relation extraction.Our model(MCTPL)Multi-granularity Convolutional Tokens Pairs of Labeling better utilizes the elements of triples and improves the ability to extract overlapping triples compared to previous approaches.Finally,we evaluated our model on two publicly accessible datasets.The experimental results show that our model’s ability to extract relation triples on the CPI dataset improves the F1 score by 2.34%compared to the current optimal model.On the DDI dataset,the F1 value improves the F1 value by 1.68%compared to the current optimal model.Our model achieved state-of-the-art performance compared to other baseline models in biomedical text entity relation extraction.展开更多
Electric vehicles use electric motors, which turn electrical energy into mechanical energy. As electric motors are conventionally used in all the industry, it is an established development site. It’s a mature technol...Electric vehicles use electric motors, which turn electrical energy into mechanical energy. As electric motors are conventionally used in all the industry, it is an established development site. It’s a mature technology with ideal power and torque curves for vehicular operation. Conventional vehicles use oil and gas as fuel or energy storage. Although they also have an excellent economic impact, the continuous use of oil and gas threatened the world’s reservation of total oil and gas. Also, they emit carbon dioxide and some toxic ingredients through the vehicle’s tailpipe, which causes the greenhouse effect and seriously impacts the environment. So, as an alternative, electric car refers to a green technology of decarbonization with zero emission of greenhouse gases through the tailpipe. So, they can remove the problem of greenhouse gas emissions and solve the world’s remaining non-renewable energy storage problem. Pure electric vehicles (PEV) can be applied in all spheres, but their special implementation can only be seen in downhole operations. They are used for low noise and less pollution in the downhole process. In this study, the basic structure of the pure electric command vehicle is studied, the main components of the command vehicle power system, namely the selection of the drive motor and the power battery, are analyzed, and the main parameters of the drive motor and the power battery are designed and calculated. The checking calculation results show that the power and transmission system developed in this paper meets the design requirements, and the design scheme is feasible and reasonable.展开更多
The present study utilized motor imaginary-based brain-computer interface technology combined with rehabilitation training in 20 stroke patients. Results from the Berg Balance Scale and the Holden Walking Classificati...The present study utilized motor imaginary-based brain-computer interface technology combined with rehabilitation training in 20 stroke patients. Results from the Berg Balance Scale and the Holden Walking Classification were significantly greater at 4 weeks after treatment （P 〈 0.01）, which suggested that motor imaginary-based brain-computer interface technology improved balance and walking in stroke patients.展开更多
The technology of remote transaction mirror image is a feasible, low-investment and well-effect disaster backup scheme in finance business system. The basic idea, construction, working principles and characteristic of...The technology of remote transaction mirror image is a feasible, low-investment and well-effect disaster backup scheme in finance business system. The basic idea, construction, working principles and characteristic of remote transaction mirror image are presented in this paper. We analyze and compare similarities and differences among this disaster backup scheme and others usually used. The technology of remote transaction mirror image have the advantages such as less requiring of software and hardware system platform, low-investment, being able to control and restore lost data, insuring the data consistency and integrity.展开更多
Dealing with data scarcity is the biggest challenge faced by Artificial Intelligence(AI),and it will be interesting to see how we overcome this obstacle in the future,but for now,“THE SHOW MUST GO ON!!!”As AI spread...Dealing with data scarcity is the biggest challenge faced by Artificial Intelligence(AI),and it will be interesting to see how we overcome this obstacle in the future,but for now,“THE SHOW MUST GO ON!!!”As AI spreads and transforms more industries,the lack of data is a significant obstacle:the best methods for teaching machines how real-world processes work.This paper explores the considerable implications of data scarcity for the AI industry,which threatens to restrict its growth and potential,and proposes plausible solutions and perspectives.In addition,this article focuses highly on different ethical considerations:privacy,consent,and non-discrimination principles during AI model developments under limited conditions.Besides,innovative technologies are investigated through the paper in aspects that need implementation by incorporating transfer learning,few-shot learning,and data augmentation to adapt models so they could fit effective use processes in low-resource settings.This thus emphasizes the need for collaborative frameworks and sound methodologies that ensure applicability and fairness,tackling the technical and ethical challenges associated with data scarcity in AI.This article also discusses prospective approaches to dealing with data scarcity,emphasizing the blend of synthetic data and traditional models and the use of advanced machine learning techniques such as transfer learning and few-shot learning.These techniques aim to enhance the flexibility and effectiveness of AI systems across various industries while ensuring sustainable AI technology development amid ongoing data scarcity.展开更多
The utilization of millimeter-wave frequencies and cognitive radio(CR)are promising ways to increase the spectral efficiency of wireless communication systems.However,conventional CR spectrum sensing techniques entail...The utilization of millimeter-wave frequencies and cognitive radio(CR)are promising ways to increase the spectral efficiency of wireless communication systems.However,conventional CR spectrum sensing techniques entail sampling the received signal at a Nyquist rate,and they are not viable for wideband signals due to their high cost.This paper expounds on how sub-Nyquist sampling in conjunction with deep learning can be leveraged to remove this limitation.To this end,we propose a multi-task learning(MTL)framework using convolutional neural networks for the joint inference of the underlying narrowband signal number,their modulation scheme,and their location in a wideband spectrum.We demonstrate the effectiveness of the proposed framework for real-world millimeter-wave wideband signals collected by physical devices,exhibiting a 91.7% accuracy in the joint inference task when considering up to two narrowband signals over a wideband spectrum.Ultimately,the proposed data-driven approach enables on-the-fly wideband spectrum sensing,combining accuracy,and computational efficiency,which are indispensable for CR and opportunistic networking.展开更多
Recent advances in artificial intelligence and the availability of large-scale benchmarks have made deepfake video generation and manipulation easier.Therefore,developing reliable and robust deepfake video detection m...Recent advances in artificial intelligence and the availability of large-scale benchmarks have made deepfake video generation and manipulation easier.Therefore,developing reliable and robust deepfake video detection mechanisms is paramount.This research introduces a novel real-time deepfake video detection framework by analyzing gaze and blink patterns,addressing the spatial-temporal challenges unique to gaze and blink anomalies using the TimeSformer and hybrid Transformer-CNN models.The TimeSformer architecture leverages spatial-temporal attention mechanisms to capture fine-grained blinking intervals and gaze direction anomalies.Compared to state-of-the-art traditional convolutional models like MesoNet and EfficientNet,which primarily focus on global facial features,our approach emphasizes localized eye-region analysis,significantly enhancing detection accuracy.We evaluate our framework on four standard datasets:FaceForensics,CelebDF-V2,DFDC,and FakeAVCeleb.The proposed framework results reveal higher accuracy,with the TimeSformer model achieving accuracies of 97.5%,96.3%,95.8%,and 97.1%,and with the hybrid Transformer-CNN model demonstrating accuracies of 92.8%,91.5%,90.9%,and 93.2%,on FaceForensics,CelebDF-V2,DFDC,and FakeAVCeleb datasets,respectively,showing robustness in distinguishing manipulated from authentic videos.Our research provides a robust state-of-the-art framework for real-time deepfake video detection.This novel study significantly contributes to video forensics,presenting scalable and accurate real-world application solutions.展开更多
Artificial intelligence(AI)is upending industries and,in many cases,supplanting traditional computer science techniques.This transition from an emerging technology to an established one is putting new burdens on unive...Artificial intelligence(AI)is upending industries and,in many cases,supplanting traditional computer science techniques.This transition from an emerging technology to an established one is putting new burdens on universities to adapt.The Fifth Global Forum on Development of Computer Science convened at Tsinghua University in the spring of this year to discuss these changes under the theme of“Development of Computer Science in the AI Era”with the aim of reaching a consensus regarding these challenges.Six heads of computer science departments shared keynotes exploring issues such as how to adapt academic and research programs to align with the advancement of AI,how to align educational programs to provide necessary AI skills,how to raise awareness around AI ethics,and how to evaluate researcher productivity.Throughout these discussions,various approaches emerged,including structurally distinguishing AI initiatives from core programs to support more targeted programs,introducing new programs that expand access to AI education,and using AI tools to support personalized education.展开更多
Large models have been widely used in the field of neural language processing,information retrieving,etc.With the development of the large models,not only is the parameter scale increased,but the model architecture ha...Large models have been widely used in the field of neural language processing,information retrieving,etc.With the development of the large models,not only is the parameter scale increased,but the model architecture has also become more complex.For example,the multi-modal transformer-based model mainly has concurrent branches,which we denoted as the concurrent branch model(CBM).Many CBMs have enlarged to tens of billions of parameters,and require distributed resources to train this kind of model.Existing distributed training systems cannot fully handle this type of model architecture because there are interactions between branches.Inspired by the unbalanced resource usage of pipeline parallelism,we prefer to organize different branches with a fine-grained bidirectional pipeline schedule of communication and computation.However,improper coordination between branches leads to idle time for computation and low training efficiency.In this paper,we present Flexpipe,a pipeline engine for c3oncurrent-branch models.We first introduce a branch-aware pipeline parallelism(BAPP)to make full use of the concurrent characteristic of the model architecture.Then,based on a multi-branch pipeline simulator,we propose an adaptive interaction coordinator,which facilitates the low-overhead branch interactions during the distributed model training.We evaluate our approach on popular concurrent branch models combined with modern training systems.Compared with the Chimera,the experiential results show that our method improves the end-to-end training throughput by 20%on average.展开更多
Software defect prediction is a critical component in maintaining software quality,enabling early identification and resolution of issues that could lead to system failures and significant financial losses.With the in...Software defect prediction is a critical component in maintaining software quality,enabling early identification and resolution of issues that could lead to system failures and significant financial losses.With the increasing reliance on user-generated content,social media reviews have emerged as a valuable source of real-time feedback,offering insights into potential software defects that traditional testing methods may overlook.However,existing models face challenges like handling imbalanced data,high computational complexity,and insufficient inte-gration of contextual information from these reviews.To overcome these limitations,this paper introduces the SESDP(Sentiment Analysis-Based Early Software Defect Prediction)model.SESDP employs a Transformer-Based Multi-Task Learning approach using Robustly Optimized Bidirectional Encoder Representations from Transformers Approach(RoBERTa)to simultaneously perform sentiment analysis and defect prediction.By integrating text embedding extraction,sentiment score computation,and feature fusion,the model effectively captures both the contextual nuances and sentiment expressed in user reviews.Experimental results show that SESDP achieves superior performance with an accuracy of 96.37%,precision of 94.7%,and recall of 95.4%,particularly excelling in handling imbalanced datasets compared to baseline models.This approach offers a scalable and efficient solution for early software defect detection,enhancing proactive software quality assurance.展开更多
Malware continues to pose a significant threat to cybersecurity,with new advanced infections that go beyond traditional detection.Limitations in existing systems include high false-positive rates,slow system response ...Malware continues to pose a significant threat to cybersecurity,with new advanced infections that go beyond traditional detection.Limitations in existing systems include high false-positive rates,slow system response times,and inability to respond quickly to new malware forms.To overcome these challenges,this paper proposes OMD-RAS:Implementing Malware Detection in an Optimized Way through Real-Time and Adaptive Security as an extensive approach,hoping to get good results towards better malware threat detection and remediation.The significant steps in the model are data collection followed by comprehensive preprocessing consisting of feature engineering and normalization.Static analysis,along with dynamic analysis,is done to capture the whole spectrum of malware behavior for the feature extraction process.The extracted processed features are given with a continuous learning mechanism to the Extreme Learning Machine model of real-time detection.This OMD-RAS trains quickly and has great accuracy,providing elite,advanced real-time detection capabilities.This approach uses continuous learning to adapt to new threats—ensuring the effectiveness of detection even as strategies used by malware may change over time.The experimental results showed that OMD-RAS performs better than the traditional approaches.For instance,the OMD-RAS model has been able to achieve an accuracy of 96.23%and massively reduce the rate of false positives across all datasets while eliciting a consistently high rate of precision and recall.The model’s adaptive learning reflected enhancements on other performance measures-for example,Matthews Correlation Coefficients and Log Loss.展开更多
BACKGROUND ANAPC1,a key regulator of the ubiquitination in tumour development,has not been thoroughly studied in hepatocellular carcinoma(HCC).AIM To elucidate the expression of ANAPC1 in HCC and its potential regulat...BACKGROUND ANAPC1,a key regulator of the ubiquitination in tumour development,has not been thoroughly studied in hepatocellular carcinoma(HCC).AIM To elucidate the expression of ANAPC1 in HCC and its potential regulatory mechanism related to ubiquitination.METHODS Bulk RNA(RNA sequencing and microarrays),immunohistochemistry(IHC)tissues,and single-cell RNA sequencing(scRNA-seq)data were integrated to comprehensively investigate ANAPC1 expression in HCC.Clustered regularly interspaced short palindromic repeats analysis was performed to assess growth in HCC cell lines following ANAPC1 knockout.Enrichment analyses were conducted to explore the functions of ANAPC1.ScRNA-seq data was used to examine the cell cycle and metabolic levels.CellChat analysis was applied to investigate the interactions between ANAPC1 and different cell types.The relationship between ANAPC1 expression and drug concentration was analyzed.RESULTS ANAPC1 messenger RNA was found to be upregulated in bulk RNA,IHC tissues samples and malignant hepatocytes.The proliferation of JHH2 cell lines was most significantly inhibited after ANAPC1 knockdown.In biological pathways,the development of HCC was found to be linked to the regulation of ubiquitin-mediated proteolysis.Additionally,scRNA-seq results indicated that highly expressed ANAPC1 was in the G2/M phase,with increased glycolysis/gluconeogenesis activity.A CellChat analysis showed that ANAPC1 was associated with the regulation of the migration inhibitory factor-(cluster of differentiation 74+C-X-C chemokine receptor type 4)pathway.Higher ANAPC1 expression correlated with stronger effects of sorafenib,dasatinib,ibrutinib,lapatinib,nilotinib and afatinib.CONCLUSION The high expression level of ANAPC1 may regulate the cell cycle and metabolic levels of HCC through the ubiquitination-related pathway,thereby promoting disease progression.展开更多
Active Disturbance Rejection Control(ADRC)possesses robust disturbance rejection capabilities,making it well-suited for longitudinal velocity control.However,the conventional Extended State Observer(ESO)in ADRC fails ...Active Disturbance Rejection Control(ADRC)possesses robust disturbance rejection capabilities,making it well-suited for longitudinal velocity control.However,the conventional Extended State Observer(ESO)in ADRC fails to fully exploit feedback from first-order and higher-order estimation errors and tracking error simultaneously,thereby diminishing the control performance of ADRC.To address this limitation,an enhanced car-following algorithm utilising ADRC is proposed,which integrates the improved ESO with a feedback controller.In comparison to the conventional ESO,the enhanced version effectively utilises multi-order estimation and tracking errors.Specifically,it enhances convergence rates by incorporating feedback from higher-order estimation errors and ensures the estimated value converges to the reference value by utilising tracking error feedback.The improved ESO significantly enhances the disturbance rejection performance of ADRC.Finally,the effectiveness of the proposed algorithm is validated through the Lyapunov approach and experiments.展开更多
Cotton fibers elongate rapidly after initiation of elongation, eventually leading to the deposit of a large amount of cellulose. To reveal features of cotton fiber cells at the fast elongation and the secondary cell w...Cotton fibers elongate rapidly after initiation of elongation, eventually leading to the deposit of a large amount of cellulose. To reveal features of cotton fiber cells at the fast elongation and the secondary cell wall synthesis stages, we compared the respective transcriptomes and metabolite profiles. Comparative analysis of transcriptomes by cDNA array identified 633 genes that were differentially regulated during fiber development. Principal component analysis （PCA） using expressed genes as variables divided fiber samples into four groups, which are diagnostic of developmental stages. Similar grouping results are also found if we use non-polar or polar metabolites as variables for PCA of developing fibers. Auxin signaling, wall-loosening and lipid metabolism are highly active during fiber elongation, whereas cellulose biosynthesis is predominant and many other metabolic pathways are downregulated at the secondary cell wall synthesis stage. Transcript and metabolite profiles and enzyme activities are consistent in demonstrating a specialization process of cotton fiber development toward cellulose synthesis. These data demonstrate that cotton fiber cell at a certain stage has its own unique feature, and developmental stages of cotton fiber cells can be distinguished by their transcript and metabolite profiles. During the secondary cell wall synthesis stage, metabolic pathways are streamed into cellulose synthesis.展开更多
With the advance of smart material science,robotics is evolving from rigid robots to soft robots.Compared to rigid robots,soft robots can safely interact with the environment,easily navigate in unstructured fields,and...With the advance of smart material science,robotics is evolving from rigid robots to soft robots.Compared to rigid robots,soft robots can safely interact with the environment,easily navigate in unstructured fields,and be minimized to operate in narrow spaces,owning to the new actuation and sensing technologies developed by the smart materials.In the review,different actuation and sensing technologies based on different smart materials are analyzed and summarized.According to the driving or feedback signals,actuators are categorized into electrically responsive actuators,thermally responsive actuators,magnetically responsive actuators,and photoresponsive actuators;sensors are categorized into resistive sensors,capacitive sensors,magnetic sensors,and optical waveguide sensors.After introducing the principle and several robotic prototypes of some typical materials in each category of the actuators and sensors.The advantages and disadvantages of the actuators and sensors are compared based on the categories,and their potential applications in robotics are also presented.展开更多
The Ecological-living-productive land(ELPL)classification system was proposed in an effort to steer China's land pattern to an ecological-centered path,with the development model shifting from a single function in...The Ecological-living-productive land(ELPL)classification system was proposed in an effort to steer China's land pattern to an ecological-centered path,with the development model shifting from a single function into more integrated multifunction land use.The focus is coordinating the man-land contradictions and developing an intensive,efficient and sustainable land use policy in an increasingly tense relationship between humans and nature.Driven by socioeconomic change and rapid population growth,many cities are undergoing urban sprawl,which involves the consumption of cropland and ecological land and threatens the ecological balance.This paper aims to quantitatively analyze the critical effects of ELPL changes on eco-environmental quality according to land use classification based on leading function of ecology,living and production from 1990 to 2015 with a case study of Xining City.Also,four future land use scenarios were simulated for 2030 using the Future Land Use Simulation(FLUS)model that couples human and natural effects.Our results show a decrease in productive land(PL)and an increase in ecological land(EL)and living land(LL)in Xining City.Forestry ecological land(FEL)covered the top largest proportion;agriculture productive land(APL)showed the greatest reduction and urban and rural living land(U-RLL)presented a dramatic increase.The eco-environmental quality improved in 1990-2010,mainly affected by the conversion of APL to FEL and GEL.However,the encroachment of U-RLL into APL,other ecological land(OEL)and FEL was the main contributor to the decline in eco-environmental quality in 2010-2015 as well as the primary reason for the increase area of lower-quality.The Harmonious Development(HD)-Scenario,characterized by a rational allocation of LL and PL and a better eco-environment,would have implications for planning and monitoring future management of ELPL,and may represent a valuable reference for local policy-makers.展开更多
文摘In the very beginning,the Computer Laboratory of the University of Cambridge was founded to provide computing service for different disciplines across the university.As computer science developed as a discipline in its own right,boundaries necessarily arose between it and other disciplines,in a way that is now often detrimental to progress.Therefore,it is necessary to reinvigorate the relationship between computer science and other academic disciplines and celebrate exploration and creativity in research.To do this,the structures of the academic department have to act as supporting scaffolding rather than barriers.Some examples are given that show the efforts being made at the University of Cambridge to approach this problem.
基金supported in part by the Key Research and Development Program of Shaanxi under Grant 2023-ZDLGY-34.
文摘Spark performs excellently in large-scale data-parallel computing and iterative processing.However,with the increase in data size and program complexity,the default scheduling strategy has difficultymeeting the demands of resource utilization and performance optimization.Scheduling strategy optimization,as a key direction for improving Spark’s execution efficiency,has attracted widespread attention.This paper first introduces the basic theories of Spark,compares several default scheduling strategies,and discusses common scheduling performance evaluation indicators and factors affecting scheduling efficiency.Subsequently,existing scheduling optimization schemes are summarized based on three scheduling modes:load characteristics,cluster characteristics,and matching of both,and representative algorithms are analyzed in terms of performance indicators and applicable scenarios,comparing the advantages and disadvantages of different scheduling modes.The article also explores in detail the integration of Spark scheduling strategies with specific application scenarios and the challenges in production environments.Finally,the limitations of the existing schemes are analyzed,and prospects are envisioned.
基金Supported by Guangxi Zhuang Autonomous Region Health Commission Scientific Research Project,No.Z-A20220521Guangxi Higher Education Undergraduate Teaching Reform Project,No.2022JGA147The National College Students’Innovation and Entrepreneurship Training Program,No.202310598042.
文摘BACKGROUND Although thymopoietin(TMPO)has been elucidated to be overexpressed in cancers,its underlying mechanisms are not yet fully understood.AIM To investigate the expression and clinical significance of TMPO in papillary thyroid carcinoma(PTC).METHODS Databases such as Gene Expression Omnibus,The Cancer Genome Atlas Proand summary receiver operating characteristic curves were plotted to evaluate diagnostic performance.A Gene Set Enrichment Analysis enrichment analysis was conducted to identify TMPO-related signaling pathways.A protein interaction network was constructed to identify hub genes.The impact of TMPO on PTC cell proliferation and the effects of its knockout were analyzed using clustered regularly interspaced short palindromic repeats(CRISPR)knockout screening and the Cancer Cell Line Encyclopedia database.RESULTS The TMPO protein was significantly overexpressed in PTC tissues,primarily localized in the cytoplasm and nuclear membrane.The mRNA level analysis showed mild overexpression of TMPO in PTC tissues,with a certain discriminatory value(area under the curve=0.66).TMPO may promote cancer through involvement in cell adhesion,focal adhesion,leukocyte migration,and multiple cancer-related signaling pathways.Additionally,CRISPR gene knockout experiments confirmed that TMPO knockout significantly inhibited the proliferation of PTC cell lines,indicating its important role in tumor growth.CONCLUSION TMPO is overexpressed in PTC and may serve as a therapeutic target and molecular biomarker for PTC.
文摘At the panel session of the 3rd Global Forum on the Development of Computer Science,attendees had an opportunity to deliberate recent issues affecting computer science departments as a result of the recent growth in the field.6 heads of university computer science departments participated in the discussions,including the moderator,Professor Andrew Yao.The first issue was how universities are managing the growing number of applicants in addition to swelling class sizes.Several approaches were suggested,including increasing faculty hiring,implementing scalable teaching tools,and working closer with other departments through degree programs that integrate computer science with other fields.The second issue was about the position and role of computer science within broader science.Participants generally agreed that all fields are increasingly relying on computer science techniques,and that effectively disseminating these techniques to others is a key to unlocking broader scientific progress.
基金supported by the National Natural Science Foundation of China(Nos.62002206 and 62202373)the open topic of the Green Development Big Data Decision-Making Key Laboratory(DM202003).
文摘Extracting valuable information frombiomedical texts is one of the current research hotspots of concern to a wide range of scholars.The biomedical corpus contains numerous complex long sentences and overlapping relational triples,making most generalized domain joint modeling methods difficult to apply effectively in this field.For a complex semantic environment in biomedical texts,in this paper,we propose a novel perspective to perform joint entity and relation extraction;existing studies divide the relation triples into several steps or modules.However,the three elements in the relation triples are interdependent and inseparable,so we regard joint extraction as a tripartite classification problem.At the same time,fromthe perspective of triple classification,we design amulti-granularity 2D convolution to refine the word pair table and better utilize the dependencies between biomedical word pairs.Finally,we use a biaffine predictor to assist in predicting the labels of word pairs for relation extraction.Our model(MCTPL)Multi-granularity Convolutional Tokens Pairs of Labeling better utilizes the elements of triples and improves the ability to extract overlapping triples compared to previous approaches.Finally,we evaluated our model on two publicly accessible datasets.The experimental results show that our model’s ability to extract relation triples on the CPI dataset improves the F1 score by 2.34%compared to the current optimal model.On the DDI dataset,the F1 value improves the F1 value by 1.68%compared to the current optimal model.Our model achieved state-of-the-art performance compared to other baseline models in biomedical text entity relation extraction.
文摘Electric vehicles use electric motors, which turn electrical energy into mechanical energy. As electric motors are conventionally used in all the industry, it is an established development site. It’s a mature technology with ideal power and torque curves for vehicular operation. Conventional vehicles use oil and gas as fuel or energy storage. Although they also have an excellent economic impact, the continuous use of oil and gas threatened the world’s reservation of total oil and gas. Also, they emit carbon dioxide and some toxic ingredients through the vehicle’s tailpipe, which causes the greenhouse effect and seriously impacts the environment. So, as an alternative, electric car refers to a green technology of decarbonization with zero emission of greenhouse gases through the tailpipe. So, they can remove the problem of greenhouse gas emissions and solve the world’s remaining non-renewable energy storage problem. Pure electric vehicles (PEV) can be applied in all spheres, but their special implementation can only be seen in downhole operations. They are used for low noise and less pollution in the downhole process. In this study, the basic structure of the pure electric command vehicle is studied, the main components of the command vehicle power system, namely the selection of the drive motor and the power battery, are analyzed, and the main parameters of the drive motor and the power battery are designed and calculated. The checking calculation results show that the power and transmission system developed in this paper meets the design requirements, and the design scheme is feasible and reasonable.
基金the National Natural Science Foundation of China,No.60970062the Shanghai Pujiang Program,No.09PJ1410200
文摘The present study utilized motor imaginary-based brain-computer interface technology combined with rehabilitation training in 20 stroke patients. Results from the Berg Balance Scale and the Holden Walking Classification were significantly greater at 4 weeks after treatment （P 〈 0.01）, which suggested that motor imaginary-based brain-computer interface technology improved balance and walking in stroke patients.
基金This work was supported by"Shu Guang"project of Shanghai Municipal Education Commission and Shanghai Education Development Foundation (No.2002SG53) and was also supported by Science and Technology Foundation of Shanghai Higher Education (No.CL200222).
文摘The technology of remote transaction mirror image is a feasible, low-investment and well-effect disaster backup scheme in finance business system. The basic idea, construction, working principles and characteristic of remote transaction mirror image are presented in this paper. We analyze and compare similarities and differences among this disaster backup scheme and others usually used. The technology of remote transaction mirror image have the advantages such as less requiring of software and hardware system platform, low-investment, being able to control and restore lost data, insuring the data consistency and integrity.
基金supported by Internal Research Support Program(IRSPG202202).
文摘Dealing with data scarcity is the biggest challenge faced by Artificial Intelligence(AI),and it will be interesting to see how we overcome this obstacle in the future,but for now,“THE SHOW MUST GO ON!!!”As AI spreads and transforms more industries,the lack of data is a significant obstacle:the best methods for teaching machines how real-world processes work.This paper explores the considerable implications of data scarcity for the AI industry,which threatens to restrict its growth and potential,and proposes plausible solutions and perspectives.In addition,this article focuses highly on different ethical considerations:privacy,consent,and non-discrimination principles during AI model developments under limited conditions.Besides,innovative technologies are investigated through the paper in aspects that need implementation by incorporating transfer learning,few-shot learning,and data augmentation to adapt models so they could fit effective use processes in low-resource settings.This thus emphasizes the need for collaborative frameworks and sound methodologies that ensure applicability and fairness,tackling the technical and ethical challenges associated with data scarcity in AI.This article also discusses prospective approaches to dealing with data scarcity,emphasizing the blend of synthetic data and traditional models and the use of advanced machine learning techniques such as transfer learning and few-shot learning.These techniques aim to enhance the flexibility and effectiveness of AI systems across various industries while ensuring sustainable AI technology development amid ongoing data scarcity.
文摘The utilization of millimeter-wave frequencies and cognitive radio(CR)are promising ways to increase the spectral efficiency of wireless communication systems.However,conventional CR spectrum sensing techniques entail sampling the received signal at a Nyquist rate,and they are not viable for wideband signals due to their high cost.This paper expounds on how sub-Nyquist sampling in conjunction with deep learning can be leveraged to remove this limitation.To this end,we propose a multi-task learning(MTL)framework using convolutional neural networks for the joint inference of the underlying narrowband signal number,their modulation scheme,and their location in a wideband spectrum.We demonstrate the effectiveness of the proposed framework for real-world millimeter-wave wideband signals collected by physical devices,exhibiting a 91.7% accuracy in the joint inference task when considering up to two narrowband signals over a wideband spectrum.Ultimately,the proposed data-driven approach enables on-the-fly wideband spectrum sensing,combining accuracy,and computational efficiency,which are indispensable for CR and opportunistic networking.
文摘Recent advances in artificial intelligence and the availability of large-scale benchmarks have made deepfake video generation and manipulation easier.Therefore,developing reliable and robust deepfake video detection mechanisms is paramount.This research introduces a novel real-time deepfake video detection framework by analyzing gaze and blink patterns,addressing the spatial-temporal challenges unique to gaze and blink anomalies using the TimeSformer and hybrid Transformer-CNN models.The TimeSformer architecture leverages spatial-temporal attention mechanisms to capture fine-grained blinking intervals and gaze direction anomalies.Compared to state-of-the-art traditional convolutional models like MesoNet and EfficientNet,which primarily focus on global facial features,our approach emphasizes localized eye-region analysis,significantly enhancing detection accuracy.We evaluate our framework on four standard datasets:FaceForensics,CelebDF-V2,DFDC,and FakeAVCeleb.The proposed framework results reveal higher accuracy,with the TimeSformer model achieving accuracies of 97.5%,96.3%,95.8%,and 97.1%,and with the hybrid Transformer-CNN model demonstrating accuracies of 92.8%,91.5%,90.9%,and 93.2%,on FaceForensics,CelebDF-V2,DFDC,and FakeAVCeleb datasets,respectively,showing robustness in distinguishing manipulated from authentic videos.Our research provides a robust state-of-the-art framework for real-time deepfake video detection.This novel study significantly contributes to video forensics,presenting scalable and accurate real-world application solutions.
文摘Artificial intelligence(AI)is upending industries and,in many cases,supplanting traditional computer science techniques.This transition from an emerging technology to an established one is putting new burdens on universities to adapt.The Fifth Global Forum on Development of Computer Science convened at Tsinghua University in the spring of this year to discuss these changes under the theme of“Development of Computer Science in the AI Era”with the aim of reaching a consensus regarding these challenges.Six heads of computer science departments shared keynotes exploring issues such as how to adapt academic and research programs to align with the advancement of AI,how to align educational programs to provide necessary AI skills,how to raise awareness around AI ethics,and how to evaluate researcher productivity.Throughout these discussions,various approaches emerged,including structurally distinguishing AI initiatives from core programs to support more targeted programs,introducing new programs that expand access to AI education,and using AI tools to support personalized education.
基金supported by the National Key R&D Program of China(No.2023YFB3001704)NSFC for Young Scientists Fund(No.62402266)NSFC for Distinguished Young Scholar(No.62225206).
文摘Large models have been widely used in the field of neural language processing,information retrieving,etc.With the development of the large models,not only is the parameter scale increased,but the model architecture has also become more complex.For example,the multi-modal transformer-based model mainly has concurrent branches,which we denoted as the concurrent branch model(CBM).Many CBMs have enlarged to tens of billions of parameters,and require distributed resources to train this kind of model.Existing distributed training systems cannot fully handle this type of model architecture because there are interactions between branches.Inspired by the unbalanced resource usage of pipeline parallelism,we prefer to organize different branches with a fine-grained bidirectional pipeline schedule of communication and computation.However,improper coordination between branches leads to idle time for computation and low training efficiency.In this paper,we present Flexpipe,a pipeline engine for c3oncurrent-branch models.We first introduce a branch-aware pipeline parallelism(BAPP)to make full use of the concurrent characteristic of the model architecture.Then,based on a multi-branch pipeline simulator,we propose an adaptive interaction coordinator,which facilitates the low-overhead branch interactions during the distributed model training.We evaluate our approach on popular concurrent branch models combined with modern training systems.Compared with the Chimera,the experiential results show that our method improves the end-to-end training throughput by 20%on average.
基金funded by a grant from the Center of Excellence in Information Assurance(CoEIA),King Saud University(KSU).
文摘Software defect prediction is a critical component in maintaining software quality,enabling early identification and resolution of issues that could lead to system failures and significant financial losses.With the increasing reliance on user-generated content,social media reviews have emerged as a valuable source of real-time feedback,offering insights into potential software defects that traditional testing methods may overlook.However,existing models face challenges like handling imbalanced data,high computational complexity,and insufficient inte-gration of contextual information from these reviews.To overcome these limitations,this paper introduces the SESDP(Sentiment Analysis-Based Early Software Defect Prediction)model.SESDP employs a Transformer-Based Multi-Task Learning approach using Robustly Optimized Bidirectional Encoder Representations from Transformers Approach(RoBERTa)to simultaneously perform sentiment analysis and defect prediction.By integrating text embedding extraction,sentiment score computation,and feature fusion,the model effectively captures both the contextual nuances and sentiment expressed in user reviews.Experimental results show that SESDP achieves superior performance with an accuracy of 96.37%,precision of 94.7%,and recall of 95.4%,particularly excelling in handling imbalanced datasets compared to baseline models.This approach offers a scalable and efficient solution for early software defect detection,enhancing proactive software quality assurance.
基金supported by a grant from the Center of Excellence in Information Assurance(CoEIA),King Saud University(KSU).
文摘Malware continues to pose a significant threat to cybersecurity,with new advanced infections that go beyond traditional detection.Limitations in existing systems include high false-positive rates,slow system response times,and inability to respond quickly to new malware forms.To overcome these challenges,this paper proposes OMD-RAS:Implementing Malware Detection in an Optimized Way through Real-Time and Adaptive Security as an extensive approach,hoping to get good results towards better malware threat detection and remediation.The significant steps in the model are data collection followed by comprehensive preprocessing consisting of feature engineering and normalization.Static analysis,along with dynamic analysis,is done to capture the whole spectrum of malware behavior for the feature extraction process.The extracted processed features are given with a continuous learning mechanism to the Extreme Learning Machine model of real-time detection.This OMD-RAS trains quickly and has great accuracy,providing elite,advanced real-time detection capabilities.This approach uses continuous learning to adapt to new threats—ensuring the effectiveness of detection even as strategies used by malware may change over time.The experimental results showed that OMD-RAS performs better than the traditional approaches.For instance,the OMD-RAS model has been able to achieve an accuracy of 96.23%and massively reduce the rate of false positives across all datasets while eliciting a consistently high rate of precision and recall.The model’s adaptive learning reflected enhancements on other performance measures-for example,Matthews Correlation Coefficients and Log Loss.
基金Co-first authors:Yu-Xing Tang 0000-0003-4382-4942Co-first authors:Wei-Zi Wu+8 种基金Corresponding author:Gang Chen,MD,Professor,Department of Pathology,The First Affiliated Hospital of Guangxi Medical University,No.6 Shuangyong Road,Nanning 530021,Guangxi Zhuang Autonomous Region,China.chengang@gxmu.edu.cn,0000-0003-2402-2987Co-corresponding authors:Yan-Ting ZhanSheng-Sheng Zhou,0000-0003-2414-460XDa-Tong Zeng,0000-0002-3338-4122Guang-Cai Zheng,0009-0001-5921-6688Rong-Quan He,0000-0002-7752-2080Di-Yuan Qin,0009-0003-3214-4762Wan-Ying Huang,0000-0002-8314-5963Yu-Lu Tang,0009-0004-0462-618X。
文摘BACKGROUND ANAPC1,a key regulator of the ubiquitination in tumour development,has not been thoroughly studied in hepatocellular carcinoma(HCC).AIM To elucidate the expression of ANAPC1 in HCC and its potential regulatory mechanism related to ubiquitination.METHODS Bulk RNA(RNA sequencing and microarrays),immunohistochemistry(IHC)tissues,and single-cell RNA sequencing(scRNA-seq)data were integrated to comprehensively investigate ANAPC1 expression in HCC.Clustered regularly interspaced short palindromic repeats analysis was performed to assess growth in HCC cell lines following ANAPC1 knockout.Enrichment analyses were conducted to explore the functions of ANAPC1.ScRNA-seq data was used to examine the cell cycle and metabolic levels.CellChat analysis was applied to investigate the interactions between ANAPC1 and different cell types.The relationship between ANAPC1 expression and drug concentration was analyzed.RESULTS ANAPC1 messenger RNA was found to be upregulated in bulk RNA,IHC tissues samples and malignant hepatocytes.The proliferation of JHH2 cell lines was most significantly inhibited after ANAPC1 knockdown.In biological pathways,the development of HCC was found to be linked to the regulation of ubiquitin-mediated proteolysis.Additionally,scRNA-seq results indicated that highly expressed ANAPC1 was in the G2/M phase,with increased glycolysis/gluconeogenesis activity.A CellChat analysis showed that ANAPC1 was associated with the regulation of the migration inhibitory factor-(cluster of differentiation 74+C-X-C chemokine receptor type 4)pathway.Higher ANAPC1 expression correlated with stronger effects of sorafenib,dasatinib,ibrutinib,lapatinib,nilotinib and afatinib.CONCLUSION The high expression level of ANAPC1 may regulate the cell cycle and metabolic levels of HCC through the ubiquitination-related pathway,thereby promoting disease progression.
基金State Key Laboratory of Intelligent Green Vehicle and Mobility,Grant/Award Number:KFY2417Science and Technology Innovation 2030-“New Generation Artificial Intelligence”Major Project,Grant/Award Number:2022ZD0116305+7 种基金State Key Laboratory of Intelligent Vehicle Safety Technology,Grant/Award Number:IVSTSKL-202402Anhui Province Natural Science Funds for Distinguished Young Scholar,Grant/Award Number:2308085J02National Natural Science Foundation of China,Grant/Award Numbers:U2013601,U20A20225Wuhu Major Scientific and Technological Achievements Engineering Project,Grant/Award Number:2021zc04CAAI-Huawei Mind Spore Open Fund,Grant/Award Number:CAAIXSJLJJ-2022-011ANatural Science Foundation of Hefei,China,Grant/Award Number:202321State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body,Grant/Award Number:32215010Wuhu Municipal Science and Technology Program,Grant/Award Number:2021hg17。
文摘Active Disturbance Rejection Control(ADRC)possesses robust disturbance rejection capabilities,making it well-suited for longitudinal velocity control.However,the conventional Extended State Observer(ESO)in ADRC fails to fully exploit feedback from first-order and higher-order estimation errors and tracking error simultaneously,thereby diminishing the control performance of ADRC.To address this limitation,an enhanced car-following algorithm utilising ADRC is proposed,which integrates the improved ESO with a feedback controller.In comparison to the conventional ESO,the enhanced version effectively utilises multi-order estimation and tracking errors.Specifically,it enhances convergence rates by incorporating feedback from higher-order estimation errors and ensures the estimated value converges to the reference value by utilising tracking error feedback.The improved ESO significantly enhances the disturbance rejection performance of ADRC.Finally,the effectiveness of the proposed algorithm is validated through the Lyapunov approach and experiments.
文摘Cotton fibers elongate rapidly after initiation of elongation, eventually leading to the deposit of a large amount of cellulose. To reveal features of cotton fiber cells at the fast elongation and the secondary cell wall synthesis stages, we compared the respective transcriptomes and metabolite profiles. Comparative analysis of transcriptomes by cDNA array identified 633 genes that were differentially regulated during fiber development. Principal component analysis （PCA） using expressed genes as variables divided fiber samples into four groups, which are diagnostic of developmental stages. Similar grouping results are also found if we use non-polar or polar metabolites as variables for PCA of developing fibers. Auxin signaling, wall-loosening and lipid metabolism are highly active during fiber elongation, whereas cellulose biosynthesis is predominant and many other metabolic pathways are downregulated at the secondary cell wall synthesis stage. Transcript and metabolite profiles and enzyme activities are consistent in demonstrating a specialization process of cotton fiber development toward cellulose synthesis. These data demonstrate that cotton fiber cell at a certain stage has its own unique feature, and developmental stages of cotton fiber cells can be distinguished by their transcript and metabolite profiles. During the secondary cell wall synthesis stage, metabolic pathways are streamed into cellulose synthesis.
基金Supported by National Key Research and Development Program of China(Grant No.2019YFB 1309800)National Natural Science Foundation of China(Grant Nos.62173197,91848206)Beijing Science&Technology Project(Grant No.Z191100008019008).
文摘With the advance of smart material science,robotics is evolving from rigid robots to soft robots.Compared to rigid robots,soft robots can safely interact with the environment,easily navigate in unstructured fields,and be minimized to operate in narrow spaces,owning to the new actuation and sensing technologies developed by the smart materials.In the review,different actuation and sensing technologies based on different smart materials are analyzed and summarized.According to the driving or feedback signals,actuators are categorized into electrically responsive actuators,thermally responsive actuators,magnetically responsive actuators,and photoresponsive actuators;sensors are categorized into resistive sensors,capacitive sensors,magnetic sensors,and optical waveguide sensors.After introducing the principle and several robotic prototypes of some typical materials in each category of the actuators and sensors.The advantages and disadvantages of the actuators and sensors are compared based on the categories,and their potential applications in robotics are also presented.
基金the support of the National Natural Science Foundation of China(No.41661038)Soft Science Research Project of Science and Technology Department of Qinghai province(No.2015-ZJ-602)
文摘The Ecological-living-productive land(ELPL)classification system was proposed in an effort to steer China's land pattern to an ecological-centered path,with the development model shifting from a single function into more integrated multifunction land use.The focus is coordinating the man-land contradictions and developing an intensive,efficient and sustainable land use policy in an increasingly tense relationship between humans and nature.Driven by socioeconomic change and rapid population growth,many cities are undergoing urban sprawl,which involves the consumption of cropland and ecological land and threatens the ecological balance.This paper aims to quantitatively analyze the critical effects of ELPL changes on eco-environmental quality according to land use classification based on leading function of ecology,living and production from 1990 to 2015 with a case study of Xining City.Also,four future land use scenarios were simulated for 2030 using the Future Land Use Simulation(FLUS)model that couples human and natural effects.Our results show a decrease in productive land(PL)and an increase in ecological land(EL)and living land(LL)in Xining City.Forestry ecological land(FEL)covered the top largest proportion;agriculture productive land(APL)showed the greatest reduction and urban and rural living land(U-RLL)presented a dramatic increase.The eco-environmental quality improved in 1990-2010,mainly affected by the conversion of APL to FEL and GEL.However,the encroachment of U-RLL into APL,other ecological land(OEL)and FEL was the main contributor to the decline in eco-environmental quality in 2010-2015 as well as the primary reason for the increase area of lower-quality.The Harmonious Development(HD)-Scenario,characterized by a rational allocation of LL and PL and a better eco-environment,would have implications for planning and monitoring future management of ELPL,and may represent a valuable reference for local policy-makers.
