教学课程资源库 更多>>

关键词： Cross modal retrieval   Semantics   Taxonomy   Computer science   Systematic literature review   Training   Statistical analysis   Space vehicles   Representation learning   Deep learning   Application and outlook   cross-modal retrieval   experimental evaluation   methodological taxonomy  

摘要： With the exponential surge in diverse multimodal data, traditional unimodal retrieval methods struggle to meet the needs of users seeking access to data across various modalities. To address this, cross-modal retrieval has emerged, enabling interaction across modalities, facilitating semantic matching, and leveraging complementarity and consistency between heterogeneous data. Although prior literature has reviewed the field of cross-modal retrieval, it suffers from numerous deficiencies in terms of timeliness, taxonomy, and comprehensiveness. This article conducts a comprehensive review of cross-modal retrieval's evolution, spanning from shallow statistical analysis techniques to vision-language pretraining (VLP) models. Commencing with a comprehensive taxonomy grounded in machine learning paradigms, mechanisms, and models, this article delves deeply into the principles and architectures underpinning existing cross-modal retrieval methods. Furthermore, it offers an overview of widely used benchmarks, metrics, and performances. Lastly, this article probes the prospects and challenges that confront contemporary cross-modal retrieval, while engaging in a discourse on potential directions for further progress in the field. To facilitate the ongoing research on cross-modal retrieval, we develop a user-friendly toolbox and an open-source repository at https://***.

关键词： Collaboration   Training   Computational modeling   Data models   Servers   Computer science   Accuracy   Sustainable development   Federated learning   Convergence   Cross-Silo federated learning (FL)   efficient collaboration   fair aggregation   utopian fairness  

摘要： Although federated learning (FL) enables collaborative training across multiple data silos in a privacy-protected manner, naively minimizing the aggregated loss to facilitate an efficient federation may compromise its fairness. Many efforts have been devoted to maintaining similar average accuracy across clients by reweighing the loss function while clients' potential contributions are largely ignored. This, however, is often detrimental since treating all clients equally will harm the interests of those clients with more contribution. To tackle this issue, we introduce utopian fairness to expound the relationship between individual earning and collaborative productivity, and propose Federated-UtoPia (FedUP), a novel FL framework that balances both efficient collaboration and fair aggregation. For the distributed collaboration, we model the training process among strategic clients as a supermodular game, which facilitates a rational incentive design through the optimal reward. As for the model aggregation, we design a weight attention mechanism to compute the fair aggregation weights by minimizing the performance bias among heterogeneous clients. Particularly, we utilize the alternating optimization theory to bridge the gap between collaboration efficiency and utopian fairness, and theoretically prove that FedUP has fair model performance with fast-rate training convergence. Extensive experiments using both synthetic and real datasets demonstrate the superiority of FedUP.

关键词： Data analysis   Cells (biology)   Task analysis   Computer science   Biology   Data mining   Clustering algorithms   Coreset   large-scale datasets   ScCoreset   single - cell RNA-seq data  

摘要： The recent boom in single-cell sequencing technologies provides valuable insights into the transcriptomes of individual cells. Through single-cell data analyses, a number of biological discoveries, such as novel cell types, developmental cell lineage trajectories, and gene regulatory networks, have been uncovered. However, the massive and increasingly accumulated single-cell datasets have also posed a seriously computational and analytical challenge for researchers. To address this issue, one typically applies dimensionality reduction approaches to reduce the large-scale datasets. However, these approaches are generally computationally infeasible for tall matrices. In addition, the downstream data analysis tasks such as clustering still take a large time complexity even on the dimension-reduced datasets. We present single-cell Coreset (scCoreset), a data summarization framework that extracts a small weighted subset of cells from a huge sparse single-cell RNA-seq data to facilitate the downstream data analysis tasks. Single-cell data analyses run on the extracted subset yield similar results to those derived from the original uncompressed data. Tests on various single-cell datasets show that scCoreset outperforms the existing data summarization approaches for common downstream tasks such as visualization and clustering. We believe that scCoreset can serve as a useful plug-in tool to improve the efficiency of current single-cell RNA-seq data analyses.

关键词： Cloning   Task analysis   Organizations   Software   Manuals   Computer science   Prototypes   Software product lines, variability management   clone management   re-engineering   framework  

摘要： Customization is a general trend in software engineering, demanding systems that support variable stakeholder requirements. Two opposing strategies are commonly used to create variants: software clone & own and software configuration with an integrated platform. Organizations often start with the former, which is cheap and agile, but does not scale. The latter scales by establishing an integrated platform that shares software assets between variants, but requires high up-front investments or risky migration processes. So, could we have a method that allows an easy transition or even combine the benefits of both strategies? We propose a method and tool that supports a truly incremental development of variant-rich systems, exploiting a spectrum between the opposing strategies. We design, formalize, and prototype a variability-management framework: the virtual platform. Virtual platform bridges clone & own and platform-oriented development. Relying on programming-language independent conceptual structures representing software assets, it offers operators for engineering and evolving a system, comprising: traditional, asset-oriented operators and novel, feature-oriented operators for incrementally adopting concepts of an integrated platform. The operators record meta-data that is exploited by other operators to support the transition. Among others, they eliminate expensive feature-location effort or the need to trace clones. A cost-and-benefit analysis of using the virtual platform to simulate the development of a real-world variant-rich system shows that it leads to benefits in terms of saved effort and time for clone detection and feature location. Furthermore, we present a user study indicating that the virtual platform effectively supports exploratory and hands-on tasks, outperforming manual development concerning correctness. We also observed that participants were significantly faster when performing typical variability management tasks using the virtual platform.

关键词： computer science   control   virtual reality   brain-computer interface   artificial intelligence   cognitive function   information integration  

摘要： Featured Application Potential applications of the work related to advanced hands-free control systems based on the interaction of BCI and *** This article examines state-of-the-art research into the impact of virtual reality (VR) on brain-computer interface (BCI) performance: how the use of virtual reality can affect brain activity and neural plasticity in ways that can improve the performance of brain-computer interfaces in IoT control, e.g., for smart home purposes. Integrating BCI with VR improves the performance of brain-computer interfaces in IoT control by providing immersive, adaptive training environments that increase signal accuracy and user control. VR offers real-time feedback and simulations that help users refine their interactions with smart home systems, making the interface more intuitive and responsive. This combination ultimately leads to greater independence, efficiency, and ease of use, especially for users with mobility issues, in managing IoT-connected devices. The integration of BCI and VR shows great potential for transformative applications ranging from neurorehabilitation and human-computer interaction to cognitive assessment and personalized therapeutic interventions for a variety of neurological and cognitive disorders. The literature review highlights the significant advances and multifaceted challenges in this rapidly evolving field. Particularly noteworthy is the emphasis on the importance of adaptive signal processing techniques, which are key to enhancing the overall control and immersion experienced by individuals in virtual environments. The value of multimodal integration, in which BCI technology is combined with complementary biosensors such as gaze tracking and motion capture, is also highlighted. The incorporation of advanced artificial intelligence (AI) techniques will revolutionize the way we approach the diagnosis and treatment of neurodegenerative conditions.

关键词： Obituaries   Carver Doris L   Seismic   Eastern Europe   Computer Science   Masters Degree   Digital Library   Central And Eastern Europe   Japan Society   Science Degree   Sense Of Humor   Editor In Chief   Amicable   Great Wall   Model For Others   Past President   IEEE Transactions   IT Professionals  

摘要： Recounts the career and contributions of Doris L. Carver.

摘要： While abstraction is critical for the transferability of automated laboratory science in (bio)chemical and materials sciences, its improper implementation is a technical debt taken against the reproducibility of experimental results. Over the decades, computer science has developed guidelines and strategies for how abstractions are captured in programming languages-particularly concerning the substitutability of implementations of abstracted ideas and the clear definition of the contexts in which abstractions are used. However, few programming languages developed for automated experiments fully leverage the wisdom learned in computer science. To achieve collaborative sharing of scientific knowledge via automated laboratories, the way that experimental protocols are codified and interpreted by machine agents must use abstractions responsibly and with reproducibility, rather than solely transferability, at its core. This Review discusses how computer science principles of abstraction can be translated to create more reproducible automation as an enabler for the acceleration of collaborative research with self-driving laboratories. Digital workflow representations in automated and autonomous chemistry laboratories can achieve transferability by using abstract concepts. However, such abstractions must abide by certain rules to ensure reproducibility. Lessons learned from computer science for responsible abstraction are translated into an automated chemistry laboratory context to guide digital workflow development towards reproducibility.

关键词： computer vision education   staged framework   computational thinking   data science   symbolic AI   neural network-based AI  

摘要： Computer vision education is increasingly important in modern technology curricula;yet, it often lacks a systematic approach integrating both theoretical concepts and practical applications. This study proposes a staged framework for computer vision education designed to progressively build learners' competencies across four levels. This study proposes a four-staged framework for computer vision education, progressively introducing concepts from basic image recognition to advanced video analysis. Validity assessments were conducted twice with 25 experts in the field of AI education and curricula. The results indicated high validity of the staged framework. Additionally, a pilot program, applying computer vision to acid-base titration activities, was implemented with 40 upper secondary school students to evaluate the effectiveness of the staged framework. The pilot program showed significant improvements in students' understanding and interest in both computer vision and scientific inquiry. This research contributes to the AI educational field by offering a structured, adaptable approach to computer vision education, integrating AI, data science, and computational thinking. It provides educators with a structured guide for implementing progressive, hands-on learning experiences in computer vision, while also highlighting areas for future research and improvement in educational methodologies.

关键词： Interdisciplinary research   Co-authorship network   Egocentric network   Professional network   Exploration strategy  

摘要： To identify collaboration trends with coauthors, this paper elaborates a theoretical framework by introducing a measure to quantify exploration of the author in joining teams of coauthors with respect to the extreme exploration possibilities. Using the clustering coefficient, we gauge the team exploration from the author-centric vista evaluating configuration values of the ego networks. This value is normalized with respect to the maximum exploration possibilities for the author facilitating us to derive a measure, viz., the team exploration score for the team exploration strategy. We further derive a dynamical version of this measure. The average profiles of the exploration strategies are compared for the authors from the USA, England, and India publishing in a rapidly growing and collaboration-extensive field, viz. artificial intelligence in computer science, in the time window from 1990 to 2020. The bibliometric data are sourced from the Clarivate Web of Science. Configuration values are evaluated in the ascending year of publications in year-long time windows to compute the team exploration score for each author. Our analysis shows that the annually averaged profiles of authors corresponding to the three countries are almost constantly increasing toward high team exploration scores. Also, in the career-averaged profiles, authors publishing more than 20 papers have mostly adopted exploratory strategies.

关键词： Location awareness   Software   History   Vectors   Computer science   Codes   Aggregates   Heterogeneous graph   change history   call relationship   fault localization   attention mechanism  

摘要： Automated software fault localization has become one of the hot spots on which researchers have focused in recent years. Existing studies have shown that learning-based techniques can effectively localize faults leveraging various information. However, there exist two problems in these techniques. The first is that they simply represent various information without caring the contribution of different information. The second is that the data imbalance problem is not considered in these techniques. Thus, their effectiveness is limited in practice. In this paper, we propose HetFL, a novel heterogeneous graph-based software fault localization technique to aggregate different information into a heterogeneous graph in which program entities and test cases are regarded as nodes, and coverage, change histories, and call relationships are viewed as edges. HetFL first extracts textual and structure information from source code as attributes of nodes and integrates them to form an attribute vector. Then, for a given node, HetFL finds its neighbor nodes based on the types of edges and aggregates corresponding neighbor nodes to form type vectors. After that, the attribute vector and all the type vectors of each node are aggregated to generate the final vector representation by an attention mechanism. Finally, we leverage a convolution neural network (CNN) to obtain the suspicious score of each method. To validate the effectiveness of HetFL, experiments are conducted on the widely used dataset Defects4J (v1.2.0). The experimental results show that HetFL can localize 217 faults within Top-1 that is 25 higher than the state-of-the-art technique DeepFL, and achieve 6.37 and 5.58 in terms of MAR and MFR which improve DeepFL by 9.0% and 5.6%, respectively. In addition, we also perform experiments on the latest version of Defects4J (v2.0.0). The experimental results show that HetFL has better performance than the baseline methods.