学术视点

学术视点

日期:2025-05-26

|  来源:【字号:

题目:Distributed Collaborative Inference System in Next-Generation Networks and Communication(下一代网络与通信中的分布式协作推理系统)

作者:C. Zhang, X. Zheng, X. Tao, C. Hu, W. Zhang and L. Zhu

来源:IEEE Transactions on Cognitive Communications and NetworkingIEEE认知通信与网络汇刊), vol. 11, no. 2, pp. 923-932.

摘要:With the rapid advancement of artificial intelligence, generative artificial intelligence (GAI) has taken a leading role in transforming data processing methods. However, the high computational demands of GAI present challenges for devices with limited resources. As we move towards the sixth generation of mobile networks (6G), the higher data rates of 6G create a need for more efficient data processing in GAI. Traditional GAI, however, shows its limitations in meeting these demands. To address these challenges, we introduce a multi-level collaborative inference system designed for next-generation networks and communication. Our proposed system features a deployment strategy that assigns models of varying sizes to devices at different network layers. Then, we design a task offloading strategy to optimise both efficiency and latency. Furthermore, a modified early exit mechanism is implemented to enhance the inference process for single models. Experimental results demonstrate that our system effectively reduces inference latency while maintaining high-quality output. Specifically, compared to existing work, our system can reduce inference time by up to 17% without sacrificing the inference accuracy.

编者译:生成式人工智能(GAI)在数据处理方法的变革中发挥了主导作用,但是,GAI的高计算需求给资源有限的设备带来了挑战。尤其是第六代移动网络(6G)更高数据速率,使得GAI需要更高效的数据处理方式。然而,传统的GAI难以应对这些挑战。因此,本文提出了一种面向下一代网络与通信的多级协作推理系统。该系统的部署策略是将不同大小的模型分配到不同网络层的设备上,随后,通过任务卸载策略,优化模型的运行效率和延迟。此外,本文还提出了一种改进的提前退出机制,以增强单个模型的推理过程。实验结果表明,该系统在保持高质量输出的同时有效降低了推理延迟。与现有工作相比,该系统可以在不牺牲推理准确性的前提下,将推理时间最多缩短17%

题目:基于生成式AI的药物重定位研究

作者:龚后武、金敏

来源:大数据, 2025, 11(02): 55-72.

摘要:针对当前药物重定位研究药物适用症预测数量固定、无法全面揭示药物潜在适应症的问题,提出了生成式AI的药物重定位模型GenDrugShifter。该模型由图注意力神经网络和Transformer Decoder模块组成,能够进行端到端的药物重定位。该模型以InChI格式表示的药物分子结构为图注意力神经网络的输入,使用自监督方法学习药物活性分子结构和药物适应症之间的潜在联系,通过自回归的方法输出药物的适应症。西药重定位实验结果表明,GenDrugShifter在预测性能上优于其他4种先进药物重定位方法。GenDrugShifter能够更全面地揭示药物潜在的适应症,具有优越性和可靠性。临床数据进一步证明了其在实际应用中的有效性。

题目:数字服务规制的政策变迁:欧盟路径与中国镜鉴

作者:李桂华、贺沛沛、黄琳

来源:信息资源管理学报, 2025, 15(2): 59-72.

摘要:本文依据政策反馈理论构建分析框架,分析欧盟数字服务规制政策内容和变革实践,梳理了欧盟数字服务规制政策“宽松责任-平衡责任-勤勉责任”的变迁演进。在这个过程中,政策变迁的规制活动受到一阶前馈的资源效应和解释效应的双重影响;而新公共政策的形成则由二阶反馈的演化效应和学习效应共同塑造。基于这一逻辑,欧盟针对数字服务规制开创了一条政策变革的路径,涵盖了政策巩固、政策学习与政策调适三大核心策略。总结欧盟数字服务规制政策经验对我国科学合理地借鉴欧盟路径具有启示意义。

题目:Towards Efficient Serverless MapReduce Computing on Cloud-Native Platforms(在云原生平台上实现高效的无服务器MapReduce计算)

作者:X. Huang, R. Gu and Y. Huang

来源:Big Data Mining and Analytics(大数据挖掘与分析), vol. 8, no. 3, pp. 575-591.

摘要:MapReduce is one of the most classic and powerful parallel computing models in the field of big data. It is still active in the big data system ecosystem and is currently evolving towards cloud-native environment. Among them, due to its elasticity and ease-to-use, Serverless computing is one of the most promising directions of cloud-native technology. To support MapReduce big data computing capabilities in a Serverless environment can give full play to Serverless's advantages. However, due to different underlying system architecture, three issues will be encountered when running MapReduce jobs in the Serverless environment. Firstly, the scheduling strategy is difficult to fully utilize the available resources. Secondly, reading Shuffle index data on cloud storage is inefficient and expensive. Thirdly, cloud storage Input/Output (I/O) request latency has a long tail effect. To solve these problems, this paper proposes three strategies with a MapReduce parallel processing framework in Serverless environment. Experimental results show that compared with cutting-edge systems, our approach shortens job execution time by 25.6% on average and reduces job execution costs by 17.3%.

编者译:MapReduce是大数据领域中最经典、最强大的并行计算模型之一,目前正在向云原生环境演变。其中,无服务器计算是云原生技术最有前景的方向之一。在无服务器环境中支持MapReduce计算,可以充分发挥无服务器的优势。然而,由于底层系统架构不同,在无服务器环境中运行MapReduce时将面临三个问题。首先,调度策略难以充分利用可用资源。其次,在云存储上读取Shuffle索引数据,效率低下且成本高昂。第三,云存储输入/输出(I/O)请求延迟存在长尾效应。为了解决这些问题,本文提出了在无服务器环境下使用MapReduce并行处理框架的三种策略。实验结果表明,与尖端系统相比,该方法将作业执行时间平均缩短了25.6%,并将作业运行成本降低了17.3%

题目:算力网络资源的统一建模及标识研究

作者:吕航、邢文娟、马小婷

来源:数据与计算发展前沿, 2025, 7(2): 12-21.

摘要:本文旨在解决算力网络资源标识的关键问题,构建可支撑资源交易与协同调度的系统性方法。首先设计资源抽象描述框架,通过属性解耦实现多维度资源表征,其次提出基于多维特征的算力资源统一建模方法,建立资源分类体系;并在此基础上构建层次化资源度量模型;进而创新性地提出算力网络资源标识的生成算法;最后面向实际应用场景,研究设计算力资源交易机制和动态优先级编排调度策略。研究成果有效解决了算力资源异构性导致的算力网络运行低效的问题,提出的标识体系为资源确权交易及协同调度提供了技术支撑,为构建智能化的算力网络生态系统提供了理论和方法基础。

题目:面向地球大数据的新型计算系统设计与实践

作者:卢莎莎、牛铁、吴璨、康乐、肖海力

来源:数据与计算发展前沿, 2025, 7(2): 40-48.

摘要:本文设计并实现了一种新型超融合架构计算系统,研发了资源聚合与作业调度、HPC计算函数等服务,实现了超级计算、云计算等多元算力在单一计算系统中的集成融合与数据共享。建成了地球大数据云服务基础平台,形成了“云+超算”协同计算服务能力,满足了科研人员按需构建个性化计算环境、利用大数据与超级计算等方法协同处理科研数据需求。


附件: