检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2022年第3期3卷 391-401页

作者： Shi, Jiachen Fan, Yi Zhou, Guoqiang Shen, Jun Nanjing University of Posts and Telecommunications College of Computer Science Nanjing210046 China Nanjing University State Key Laboratory for Novel Software Technology Nanjing210046 China University of Wollongong School of Computing and Information Technology WollongongNSW2522 Australia

In the existing reinforcement learning (RL)-based neural architecture search (NAS) methods for a generative adversarial network (GAN), both the generator and the discriminator architecture are usually treated as the search objects. In this article, we take a different perspective to propose an approach by treating the generator as the search objective and the discriminator as the judge to evaluate the performance of the generator architecture. Consequently, we can convert this NAS problem to a GAN-style problem, similar to using a controller to generate sequential data via reinforcement learning in a sequence GAN, except that the controller in our methods generates serialized data information of architecture. Furthermore, we adopt an RL-based distributed search method to update the controller parameters θ. Generally, the reward value is calculated after the whole architecture searched, but as another novelty in this article, we employ the reward shaping method to judge the intermediate reward and assign it to every cell in the architecture to encourage the diversity and the integrity of all cells. The main contribution of this article is to provide a novel performance estimation mechanism, which could speed up the efficiency of architecture search, and improve the searching results with specific supplementary strategies. Crucially, this estimation mechanism can be applied to most RL-based NAS methods for the GAN. The experiments demonstrate that our methods achieve satisfactory results against our design objectives. © 2020 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Incorporating metapath interaction on heterogeneous information network for social recommendation

引用

Frontiers of computer science 2024年第1期18卷 33-48页

作者： Yanbin JIANG Huifang MA Xiaohui ZHANG Zhixin LI Liang CHANG College of Computer Science and Engineering Northwest Normal UniversityLanzhou 730070China Guangxi Key Lab of Multi-source Information Mining and Security Guangxi Normal UniversityGuilin 541001China Guangxi Key Laboratory of Trusted Software Guilin University of Electronic TechnologyGuilin 541004China

Heterogeneous information network(HIN)has recently been widely adopted to describe complex graph structure in recommendation systems,proving its effectiveness in modeling complex graph *** existing HIN-based recommendation studies have achieved great success by performing message propagation between connected nodes on the defined metapaths,they have the following major *** works mainly convert heterogeneous graphs into homogeneous graphs via defining metapaths,which are not expressive enough to capture more complicated dependency relationships involved on the ***,the heterogeneous information is more likely to be provided by item attributes while social relations between users are not adequately *** tackle these limitations,we propose a novel social recommendation model MPISR,which models MetaPath Interaction for Social Recommendation on heterogeneous information ***,our model first learns the initial node representation through a pretraining module,and then identifies potential social friends and item relations based on their similarity to construct a unified *** then develop the two-way encoder module with similarity encoder and instance encoder to capture the similarity collaborative signals and relational dependency on different *** experiments on five real datasets demonstrate the effectiveness of our method.

关键词： heterogeneous information network social reco-mmender system metapath interaction attention mechanism

来源：评论

学校读者我要写书评

暂无评论

A Heuristic-based Dynamic Scheduling and Routing Method for Industrial TSN Networks 10

A Heuristic-based Dynamic Scheduling and Routing Method for ...

引用

10th IEEE International Conference on Cyber Security and Cloud Computing and 9th IEEE International Conference on Edge Computing and Scalable Cloud, CSCloud-EdgeCom 2023

作者： Chen, Honglong Liu, Mindong Huang, Jing Zheng, Zhiling Huang, Weihong Xiao, Yufeng Hunan University of Science and Technology Hunan Key Laboratory for Service Computing and Novel Software Technology School of Computer Science and Engineering Xiangtan411201 China Hunan Industry Polytechnic Information Center Changsha410125 China

ISBN: (纸本)9798350312461

In the industrial environment, machines often need to reflect the anomaly detection results to the total control center in time, and the general industrial network can not achieve high real-time. In order to solve such challenges, a set of protocol standards developed by IEEE802.1 working group, namely Time-sensitive Networking (TSN), has been introduced into industrial networks. TSN can provide high real-time and reliability for data transmission, where the reliability is achieved by Frame duplication and Frame Elimination (FRER). In the realization process of FRER, it is necessary to determine the source node, destination node, and multiple disjoint paths to transmit redundant data. However, the transmission of these redundant traffic may result in the delay of other flows, and then affects the user experience. Therefore, it is very important to choose excellent redundant traffic paths to ensure reliability and reduce the impact on other flows. In the existing research, there are many dynamic scheduling and routing heuristics to determine the path, but they do not consider the influence of the location of the source node on the whole route scheduling. This paper proposes an improved dynamic scheduling and routing heuristic method, which takes the source node into account in the routing selection. In the flow test experiments of different magnitudes, it is found that the total delay of all flows is reduced by 1.4%-4.5% under the same magnitude of schedulability compared with Ant Colony Optimization. © 2023 IEEE.

关键词： Heuristic algorithms

来源：评论

学校读者我要写书评

暂无评论

Vision-Language Models are Strong Noisy Label Detectors 38

Vision-Language Models are Strong Noisy Label Detectors

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Wei, Tong Li, Hao-Tian Li, Chun-Shu Shi, Jiang-Xin Li, Yu-Feng Zhang, Min-Ling School of Computer Science and Engineering Southeast University Nanjing China Ministry of Education China National Key Laboratory for Novel Software Technology Nanjing University China School of Artificial Intelligence Nanjing University China

Recent research on fine-tuning vision-language models has demonstrated impressive performance in various downstream tasks. However, the challenge of obtaining accurately labeled data in real-world applications poses a significant obstacle during the fine-tuning process. To address this challenge, this paper presents a Denoising Fine-Tuning framework, called DEFT, for adapting vision-language models. DEFT utilizes the robust alignment of textual and visual features pretrained on millions of auxiliary image-text pairs to sieve out noisy labels. The proposed framework establishes a noisy label detector by learning positive and negative textual prompts for each class. The positive prompt seeks to reveal distinctive features of the class, while the negative prompt serves as a learnable threshold for separating clean and noisy samples. We employ parameter-efficient fine-tuning for the adaptation of a pre-trained visual encoder to promote its alignment with the learned textual prompts. As a general framework, DEFT can seamlessly fine-tune many pre-trained models to downstream tasks by utilizing carefully selected clean samples. Experimental results on seven synthetic and real-world noisy datasets validate the effectiveness of DEFT in both noisy label detection and image classification tasks. Our source code is available at https://***/HotanLee/DeFT. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Arctic sea ice volume export through the Fram Strait: variation and its effect factors

引用

Acta Oceanologica Sinica 2023年第5期42卷 166-178页

作者： Haili Li Changqing Ke Qinghui Zhu Xiaoyi Shen Jiangsu Provincial Key Laboratory of Geographic Information Science and Technology/Key Laboratory for Land Satellite Remote Sensing Applications of Ministry of Natural Resources School of Geography and Ocean ScienceNanjing UniversityNanjing 210023China Collaborative Innovation Center of Novel Software Technology and Industrialization Nanjing 210023China Collaborative Innovation Center of South China Sea Studies Nanjing 210023China

Arctic sea ice export is important for the redistribution of freshwater and sea ice ***,we use the sea ice thickness,sea ice velocity,and sea ice concentration(SIC)to estimate the exported sea ice volume through the Fram Strait from 2011 to *** further analyse the contributions of the sea ice thickness,velocity and concentration to sea ice volume ***,the relationships between atmospheric circulation indices(Arctic Oscillation(AO),North Atlantic Oscillation(NAO),and Arctic Dipole(AD))and the sea ice volume export are ***,we analyse the impact of wind-driven oceanic circulation indices(Ekman transport(ET))on the sea ice volume *** sea ice volume export rapidly increases in winter and decreases in *** exported sea ice volume in winter is likely to exceed that in spring in the *** sea ice thickness,velocity and SIC,the greatest contribution to sea ice export comes from the ice *** exported sea ice volume through the zonal gate of the Fram Strait(which contributes 97%to the total sea ice volume export of the Fram Strait)is much higher than that through the meridional gate(3%)because the sea ice flowing out of the zonal gate has the characteristics of a high thickness(mainly thicker than 1 m),a high velocity(mainly faster than 0.06 m/s)and a high concentration(mainly higher than 80%).The AD and ET explain 53.86%and 38.37%of the variation in sea ice volume export,respectively.

关键词： sea ice thickness sea ice velocity sea ice concentration sea ice volume export Arctic Dipole Ekman transport Fram Strait

来源：评论

学校读者我要写书评

暂无评论

AI Computing Systems for Large Language Models Training

引用

计算机科学技术学报（英文版） 2025年第1期40卷 6-41页

作者：张振兴文渊博吕涵祺刘畅张蕊李夏青王超杜子东郭崎李玲周学海陈云霁 1School of Computer Science and Technology University of Science and Technology of China Hefei 230026 China 2State Key Laboratory of Processors Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China 3Cambricon Technologies Beijing 100191 China 4Shanghai Innovation Center for Processor Technologies Shanghai 201210 China 5Intelligent Software Research Center Institute of Software Chinese Academy of Sciences Beijing 100190 China 6University of Chinese Academy of Sciences Beijing 101408 China

In this paper,we present a comprehensive overview of artificial intelligence(AI)computing systems for large language models(LLMs)*** rapid advancement of LLMs in recent years,coupled with the widespread adoption of algorithms and applications such as BERT,ChatGPT,and DeepSeek,has sparked significant interest in this *** classify LLMs into encoder-only,encoder-decoder,and decoder-only models,and briefly analyze their training and infer-ence processes to emphasize their substantial need for computational *** operations depend heavily on AI-specific accelerators like GPUs(graphics processing units),TPUs(tensor processing units),and MLUs(machine learning units).However,as the gap widens between the increasing complexity of LLMs and the current capabilities of accelerators,it becomes essential to adopt heterogeneous computing systems optimized for distributed environments to manage the growing computational and memory requirements of *** delve into the execution and scheduling of LLM algo-rithms,underlining the critical role of distributed computing strategies,memory management enhancements,and boosting computational *** paper clarifies the complex relationship between algorithm design,hardware infrastructure,and software optimization,and provides an in-depth understanding of both the software and hardware infrastructure sup-porting LLMs training,offering insights into the challenges and potential avenues for future development and deployment.

关键词： artificial intelligence(AI)chip large language model(LLM) AI computing system accelerator

来源：评论

学校读者我要写书评

暂无评论

Cantor: Inspiring Multimodal Chain-of-Thought of MLLM 24

Cantor: Inspiring Multimodal Chain-of-Thought of MLLM

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Gao, Timin Chen, Peixian Zhang, Mengdan Fu, Chaoyou Shen, Yunhang Zhang, Yan Zhang, Shengchuan Zheng, Xiawu Sun, Xing Cao, Liujuan Ji, Rongrong Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China Xiamen University Xiamen China Tencent Youtu Lab Shanghai China State Key Laboratory for Novel Software Technology Nanjing University China School of Intelligence Science and Technology Nanjing University China

ISBN: (纸本)9798400706868

With the advent of large language models(LLMs) enhanced by the chain-of-thought(CoT) methodology, the visual reasoning problem is usually decomposed into manageable sub-tasks and tackled sequentially with various external tools. However, such a paradigm faces the challenge of the potential "determining hallucinations"in decision generation due to insufficient visual information and the limitation of low-level perception tools that fail to provide abstract summaries necessary for comprehensive reasoning. We argue that converging visual context acquisition and logical reasoning is pivotal for tackling visual reasoning tasks. This paper delves into the realm of multimodal CoT to solve intricate visual reasoning tasks with multimodal large language models(MLLMs) and their cognitive capability. To this end, we propose an innovative multimodal CoT framework, termed Cantor, characterized by a perception-decision architecture. Cantor first acts as a decision generator and integrates visual inputs to analyze the image and problem, ensuring a closer alignment with the actual context. Furthermore, Cantor leverages the advanced cognitive functions of MLLMs to perform as multifaceted experts for deriving higher-level information, enhancing the CoT generation process. Our extensive experiments demonstrate the efficacy of the proposed framework, showing significant improvements in multimodal CoT performance across two complex visual reasoning datasets, without necessitating fine-tuning or ground-truth rationales. Project Page: https://***/cantor/. © 2024 ACM.

关键词： Chains

来源：评论

学校读者我要写书评

暂无评论

Buffered Asynchronous Secure Aggregation for Cross-Device Federated Learning

arXiv

引用

arXiv 2024年

作者： Wang, Kun Yang, Yi-Rui Li, Wu-Jun National Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing University China

Asynchronous federated learning (AFL) is an effective method to address the challenge of device heterogeneity in cross-device federated learning. However, AFL is usually incompatible with existing secure aggregation protocols used to protect user privacy in federated learning because most existing secure aggregation protocols are based on synchronous aggregation. To address this problem, we propose a novel secure aggregation protocol named buffered asynchronous secure aggregation (BASA) in this paper. Compared with existing protocols, BASA is fully compatible with AFL and provides secure aggregation under the condition that each user only needs one round of communication with the server without relying on any synchronous interaction among users. Based on BASA, we propose the first AFL method which achieves secure aggregation without extra requirements on hardware. We empirically demonstrate that BASA outperforms existing secure aggregation protocols for cross-device federated learning in terms of training efficiency and scalability. Copyright © 2024, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

A quantum federated learning framework for classical clients

引用

science China(Physics,Mechanics & Astronomy) 2024年第5期67卷 1-10页

作者： Yanqi Song Yusen Wu Shengyao Wu Dandan Li Qiaoyan Wen Sujuan Qin Fei Gao State Key Laboratory of Networking and Switching Technology Beijing University of Posts and TelecommunicationsBeijing 00876China Department of Physics The University of Western AustraliaPerthWA 6009Australia School of Computer Science(National Pilot Software Engineering School) Beijing University of Posts and TelecommunicationsBeijing100876China

Quantum federated learning(QFL)enables collaborative training of a quantum machine learning(QML)model among multiple clients possessing quantum computing capabilities,without the need to share their respective local ***,the limited availability of quantum computing resources poses a challenge for each client to acquire quantum computing *** raises a natural question:Can quantum computing capabilities be deployed on the server instead?In this paper,we propose a QFL framework specifically designed for classical clients,referred to as CC-QFL,in response to this *** each iteration,the collaborative training of the QML model is assisted by the shadow tomography technique,eliminating the need for quantum computing capabilities of ***,the server constructs a classical representation of the QML model and transmits it to the *** clients encode their local data onto observables and use this classical representation to calculate local *** local gradients are then utilized to update the parameters of the QML *** evaluate the effectiveness of our framework through extensive numerical simulations using handwritten digit images from the MNIST *** framework provides valuable insights into QFL,particularly in scenarios where quantum computing resources are scarce.

关键词： client server quantum

来源：评论

学校读者我要写书评

暂无评论

Accelerated glacier mass loss in the southeastern Tibetan Plateau since the 1970s

引用

Advances in Climate Change Research 2023年第3期14卷 372-386页

作者： Lan-Hua LUO Chang-Qing KE Yu-Bin FAN Zi-Fei WANG Jiangsu Provincial Key Laboratory of Geographic Information Science and Technology Key Laboratory for Land Satellite Remote Sensing Applications of Ministry of Natural ResourcesSchool of Geography and Ocean ScienceNanjing UniversityNanjing 210023China Collaborative Innovation Center of Novel Software Technology and Industrialization Nanjing 210023China Collaborative Innovation Center of South China Sea Studies Nanjing 210023China

The southeastern Tibetan Plateau(SETP)is a region in High Mountain Asia with the most serious glacier mass ***,long-term and large regional-scale studies that estimate glacier mass balance in this area remain *** this study,we generated a KH-9 Digital Elevation Model(DEM)(covering 87.6%of the glacier area)for the 1970s from KH-9 Hexagon imagery and quantified geodetic glacier mass over the SETP from the 1970s to 2020 using KH-9 DEM,NASADEM and ICESat-2 ATL06 *** results show that the SETP was in a state of serious mass loss(at a rate of−0.35±0.03 m *** year)from the 1970s to 2020,and the rate of mass loss accelerated from−0.15±0.05 to−0.56±0.10 m *** year between the 1970s–2000 and 2000–*** the SETP,the glacier mass balance revealed a remarkable spatial *** maximum mean glacier mass loss rate was observed in the Hengduan *** exhibited the highest acceleration in the mass loss rate since the *** warming of air temperature and decreasing snowfall can partly explain the accelerated glacier mass loss in the *** study provides a new long-term glacier mass balance estimation covering almost the entire SETP that suggests the acceleration in glacier mass loss observed in the SETP since the 1970s is a regional tendency,which is crucial for understanding the relationship between glaciers and climate changes.

关键词： Glacier mass balance KH-9 hexagon imagery NASADEM ICESat-2 Southeastern Tibetan Plateau

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：