检索结果-内蒙古大学图书馆

Large circuit models: opportunities and challenges

Science China(Information Sciences) 2024年第10期67卷 25-66页

作者： Lei CHEN Yiqi CHEN Zhufei CHU Wenji FANG Tsung-Yi HO Ru HUANG Yu HUANG Sadaf KHAN Min LI Xingquan LI Yu LI Yun LIANG Jinwei LIU Yi LIU Yibo LIN Guojie LUO Hongyang PAN Zhengyuan SHI Guangyu SUN Dimitrios TSARAS Runsheng WANG Ziyi WANG Xinming WEI Zhiyao XIE Qiang XU Chenhao XUE Junchi YAN Jun YANG Bei YU Mingxuan YUAN Evangeline F.Y.YOUNG Xuan ZENG Haoyi ZHANG Zuodong ZHANG Yuxiang ZHAO Hui-Ling ZHEN Ziyang ZHENG Binwu ZHU Keren ZHU Sunan ZOU Huawei Noah’s Ark Lab School of Integrated Circuits Peking University Faculty of Electrical Engineering and Computer Science Ningbo University Department of Electronic and Computer Engineering Hong Kong University of Science and Technology Department of Computer Science and Engineering The Chinese University of Hong Kong School of Integrated Circuits Southeast University Huawei HiSilicon Peng Cheng Laboratory School of Computer Science Peking University School of Microelectronics State Key Laboratory of Integrated Chips and System Fudan University School of Artificial Intelligence Shanghai Jiao Tong University

Within the electronic design automation(EDA) domain, artificial intelligence(AI)-driven solutions have emerged as formidable tools, yet they typically augment rather than redefine existing methodologies. These solutions often repurpose deep learning models from other domains, such as vision, text, and graph analytics, applying them to circuit design without tailoring to the unique complexities of electronic circuits. Such an “AI4EDA” approach falls short of achieving a holistic design synthesis and understanding,overlooking the intricate interplay of electrical, logical, and physical facets of circuit data. This study argues for a paradigm shift from AI4EDA towards AI-rooted EDA from the ground up, integrating AI at the core of the design process. Pivotal to this vision is the development of a multimodal circuit representation learning technique, poised to provide a comprehensive understanding by harmonizing and extracting insights from varied data sources, such as functional specifications, register-transfer level(RTL) designs, circuit netlists,and physical layouts. We champion the creation of large circuit models(LCMs) that are inherently multimodal, crafted to decode and express the rich semantics and structures of circuit data, thus fostering more resilient, efficient, and inventive design methodologies. Embracing this AI-rooted philosophy, we foresee a trajectory that transcends the current innovation plateau in EDA, igniting a profound “shift-left” in electronic design methodology. The envisioned advancements herald not just an evolution of existing EDA tools but a revolution, giving rise to novel instruments of design-tools that promise to radically enhance design productivity and inaugurate a new epoch where the optimization of circuit performance, power, and area(PPA) is achieved not incrementally, but through leaps that redefine the benchmarks of electronic systems' capabilities.

关键词： AI-rooted EDA large circuit models (LCMs) multimodal circuit representation learning circuit optimization

来源：评论

学校读者我要写书评

暂无评论

Learning to Branch in Combinatorial Optimization With Graph Pointer Networks

引用

IEEE/CAA Journal of Automatica Sinica 2024年第1期11卷 157-169页

作者： Rui Wang Zhiming Zhou Kaiwen Li Tao Zhang Ling Wang Xin Xu Xiangke Liao IEEE the Xiangjiang Laboratory and the College of Systems Engineering National University of Defense TechnologyChangsha 410073China the Institute of Automation Chinese Academy of SciencesBeijing 100190China the College of Systems Engineering National University of Defense TechnologyChangsha 410073 the Hunan Key Laboratory of Multi-Energy System Intelligent Interconnection Technology Changsha 410073China the Department of Automation Tsinghua UniversityBeijing 100084China the College of Intelligence Science and Technology National University of Defense TechnologyChangsha 410073China the College of Computer Science and Technology National University of Defense TechnologyChangsha 410073China

Traditional expert-designed branching rules in branch-and-bound(B&B) are static, often failing to adapt to diverse and evolving problem instances. Crafting these rules is labor-intensive, and may not scale well with complex *** the frequent need to solve varied combinatorial optimization problems, leveraging statistical learning to auto-tune B&B algorithms for specific problem classes becomes attractive. This paper proposes a graph pointer network model to learn the branch rules. Graph features, global features and historical features are designated to represent the solver state. The graph neural network processes graph features, while the pointer mechanism assimilates the global and historical features to finally determine the variable on which to branch. The model is trained to imitate the expert strong branching rule by a tailored top-k Kullback-Leibler divergence loss function. Experiments on a series of benchmark problems demonstrate that the proposed approach significantly outperforms the widely used expert-designed branching rules. It also outperforms state-of-the-art machine-learning-based branch-and-bound methods in terms of solving speed and search tree size on all the test instances. In addition, the model can generalize to unseen instances and scale to larger instances.

关键词： Branch-and-bound(B&B) combinatorial optimization deep learning graph neural network imitation learning

来源：评论

学校读者我要写书评

暂无评论

Compression Format and Systolic Array Structure Co-design for Accelerating Sparse Matrix Multiplication in DNNs 24th

Compression Format and Systolic Array Structure Co-design f...

引用

24th International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2024

作者： Cao, Yongxiang Jiang, Jixiang Zhao, Guocheng Wang, Wei Jiang, Hongxu Song, Yanfei School of Computer Science and Engineering Beihang University Beijing China Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing China

ISBN: (纸本)9789819615445

The systolic array (SA) architecture is widely used in accelerator/AI chip design due to its excellent matrix multiplication acceleration effect. However, the mismatch between the traditional sparse matrix compression format and the SA architecture results in poor SA performance in computing sparse matrices. To solve the problem, we propose a method for co-designing the sparse compression format and SA architecture. This work proposes the Vector Group Compressed Coordinate (VGCC) data stream compression format and designs a sparse SA architecture based on hardware encoding and decoding. VGCC can handle data streams of any size without increasing the index bit width. The VGCC compression size is significantly smaller than general sparse compression formats such as COO/CSR. In a sparse matrix with a sparsity of 90%, it saves 67.13% of storage space compared to the state-of-the-art ECOO data stream compression format and achieves an average acceleration of 1.54 times. We also propose a VGCC-friendly SA architecture. Our SA architecture engine reduces logic consumption by 51.6% compared to Sparse TPU while achieving a 3.23x speedup. To mitigate the issue of high no-load rates in systolic arrays, we introduce a Data Transmission Priority (DTP) data matching algorithm. With sparsity levels ranging from 50% to 90%, the algorithm’s calculation speed is enhanced by 1.124x to 1.991x. We implemented a sparse DNN accelerator on an FPGA Zynq UltraScale+MPSoC ZCU102, utilizing the VGCC compression format proposed in this study for compressing input data. When compared to the S2 Engine, the accelerator achieved a 3.37x speed improvement in matrix multiplication with 75% sparsity, significantly enhancing the speed of sparse matrix multiplication. The accelerator achieved an average throughput of 1079.33GOP/s when inferencing DNNs. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Systolic arrays

来源：评论

学校读者我要写书评

暂无评论

A multi-channel image-based method for flight delay prediction in terminal area 4

A multi-channel image-based method for flight delay predicti...

引用

2024 4th International Conference on Intelligent Traffic systems and Smart City, ITSSC 2024

作者： Yang, Yihan Chen, Haiyan Yang, Longwu State Key Laboratory of Air Traffic Management System Nanjing China College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing China

ISBN: (纸本)9781510686304

With the rapid development of civil aviation, flight delays have become increasingly critical. Effectively predicting flight delays has become a research hotspot in recent years. In this paper, we propose a novel flight delay prediction method based on image recognition by applying deep learning techniques. We first rasterize the target terminal area into a grid image of size 25×25. Then, we convert the arrival and departure traffic flow data collected over some time into colors and fill them into the corresponding grids to obtain the arrival and departure traffic flow images. On this basis, we design the prediction model Attention-Complementary Hybrid Network (ACHNet), which can be used for multi-channel image feature fusion. It takes arrival and departure traffic flow images as inputs and fuses the spatial information of these two types of images through the fusion branch. We use the Convolutional Neural Network (CNN)-based prediction model to predict flight delays on arrival and departure traffic flow image sets respectively, and compare the prediction results with those of ACHNet. The experimental results show that the fusion of multi-channel image features can effectively improve the performance of flight delay prediction. © 2025 SPIE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Recent Progress in Reinforcement Learning and Adaptive Dynamic Programming for Advanced Control Applications

引用

IEEE/CAA Journal of Automatica Sinica 2024年第1期11卷 18-36页

作者： Ding Wang Ning Gao Derong Liu Jinna Li Frank L.Lewis IEEE the Faculty of Information Technology Beijing Key Laboratory of Computational Intelligence and Intelligent SystemBeijing Laboratory of Smart Environmental Protectionand Beijing Institute of Artificial IntelligenceBeijing University of TechnologyBeijing 100124China the School of System Design and Intelligent Manufacturing Southern University of Science and TechnologyShenzhen 518055China the Department of Electrical and Computer Engineering University of Illinois at ChicagoChicago IL 60607 USA the School of Information and Control Engineering Liaoning Petrochemical UniversityFushun 113001China the UTA Research Institute the University of Texas at ArlingtonArlington TX 76118 USA

Reinforcement learning(RL) has roots in dynamic programming and it is called adaptive/approximate dynamic programming(ADP) within the control community. This paper reviews recent developments in ADP along with RL and its applications to various advanced control fields. First, the background of the development of ADP is described, emphasizing the significance of regulation and tracking control problems. Some effective offline and online algorithms for ADP/adaptive critic control are displayed, where the main results towards discrete-time systems and continuous-time systems are surveyed, ***, the research progress on adaptive critic control based on the event-triggered framework and under uncertain environment is discussed, respectively, where event-based design, robust stabilization, and game design are reviewed. Moreover, the extensions of ADP for addressing control problems under complex environment attract enormous attention. The ADP architecture is revisited under the perspective of data-driven and RL frameworks,showing how they promote ADP formulation ***, several typical control applications with respect to RL and ADP are summarized, particularly in the fields of wastewater treatment processes and power systems, followed by some general prospects for future research. Overall, the comprehensive survey on ADP and RL for advanced control applications has d emonstrated its remarkable potential within the artificial intelligence era. In addition, it also plays a vital role in promoting environmental protection and industrial intelligence.

关键词： Adaptive dynamic programming(ADP) advanced control complex environment data-driven control event-triggered design intelligent control neural networks nonlinear systems optimal control reinforcement learning(RL)

来源：评论

学校读者我要写书评

暂无评论

CFinDEE: A Chinese Fine-Grained Financial Dataset for Document-Level Event Extraction 24

CFinDEE: A Chinese Fine-Grained Financial Dataset for Docume...

引用

33rd Companion of the ACM World Wide Web Conference, WWW 2023

作者： Zhang, Tian Liu, Maofu Zhou, Bingying School of Computer Science and Technology Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan University of Science and Technology Wuhan China

ISBN: (纸本)9798400701726

Document-level event extraction faces numerous challenges in accurately modeling real-world financial scenarios, particularly due to the inadequacies in existing datasets regarding data scale and fine-grained annotations. The development of datasets is a crucial factor in driving research progress;therefore, we present a high-quality Chinese document-level event extraction dataset, CFinDEE. This dataset, grounded in real-world financial news, defines 22 event types and 116 argument roles, annotating 26,483 events and 107,096 event arguments. CFinDEE aims to address these shortcomings by providing more comprehensive annotations and data augmentation, offering richer resources for document-level event extraction in the financial domain. CFinDEE extends data both horizontally and vertically, where horizontal expansion enriches the types of financial events, enhancing the diversity of the dataset;vertical expansion, by increasing the scale of the data, effectively boosts the practical value of the dataset. Experiments conducted on multiple advanced models have validated the high applicability and effectiveness of the CFinDEE dataset for document-level event extraction tasks in the financial field. © 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Expansion

来源：评论

学校读者我要写书评

暂无评论

Adaptive Granulation for Efficient Classification on Noise Labeled Images 9

Adaptive Granulation for Efficient Classification on Noise L...

引用

9th IEEE International Conference on Computational Intelligence and Applications, ICCIA 2024

作者： Cai, Kecan Zhang, Hongyun Miao, Duoqian Tongji University Department of Computer Science and Technology Shanghai China Key Laboratory of Embedded System and Service Computing Ministry of Education Shanghai China

ISBN: (纸本)9798350352214

Current semi-supervised learning-based sample selection methods for noisy label image classification typically utilize all clean and noisy samples for model training. However, not all noisy samples contribute positively to model training. This paper introduces a novel semi-supervised image data granulation method that employs adaptively generated granular noisy sample subsets in place of the original noisy samples to enhance classification efficiency. The granular data generated retain the essential features of the original data, thereby improving efficiency without compromising classification accuracy. The quality of the granular data is assessed using coverage and specificity criteria, standard metrics for evaluating information granules. The proposed method consists of three main components: (i)selecting clean and noisy samples through network co-training, (ii)calculating granular noisy sample subsets by adaptive granulation, and (iii)optimizing the network model using a semi-supervised strategy. Experimental results on benchmark datasets with varying noise rates demonstrate that our method significantly improves the efficiency of noisy label image classification while maintaining accuracy. © 2024 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Unknown Task Selection and Worker Recruitment Using Two-Stage Multi-Armed Bandit in Crowdsensing

引用

IEEE Internet of Things Journal 2025年第13期12卷 24441-24456页

作者： Song, Haoyuan Li, Peng Du, Rui Yu, Hai Huang, Weiyi Nie, Lei Bao, Haizhou Liu, Qin Wuhan University of Science and Technology School of Computer Science and Technology Wuhan430065 China Wuhan University of Science and Technology Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan430065 China Hubei Polytechnic University School of Computer Huangshi435003 China Jilin University Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Changchun130012 China Wuhan University School of Cyber Science and Engineering Wuhan430072 China

Mobile Crowdsensing (MCS) faces significant challenges in selecting tasks with expected high revenue and recruiting workers with expected high qualities to maximize overall utility. Existing approaches often assume the revenue of posting tasks or the quality of workers is determined, which limits their practical applicability. This paper tackles these challenges by modeling the long-term multi-task, multi-worker selection problem as a Two-Stage Multi-Armed Bandit (TS-MAB) problem under uncertainty. In each round, the MCS platform selects a long-term task from a pool and recruits the necessary workers, determining payments accordingly. We model the task selection phase as a MAB problem, and the worker recruitment phase, influenced by the task selection, as a Combinatorial MAB (CMAB) problem. We propose an extended Upper Confidence Bound (UCB)-based strategy and develop an incentive mechanism based on Auction theory, combined with TS-MAB (i.e., ATS-MAB), for unknown task selection and worker recruitment. Our mechanism theoretically guarantees truthfulness and individual rationality, with a theoretical analysis of its regret bounds. Furthermore, we introduce an adaptive incentive mechanism called AATS-MAB, which improves worker recruitment and quality updates, achieving higher total sensing quality and lower regret. Extensive simulations demonstrate the effectiveness and scalability of the proposed methods. © 2014 IEEE.

关键词： Personnel selection

来源：评论

学校读者我要写书评

暂无评论

Super-resolution reconstruction of single image for latent features

引用

Computational Visual Media 2024年第6期10卷 1219-1239页

作者： Xin Wang Jing-Ke Yan Jing-Ye Cai Jian-Hua Deng Qin Qin Yao Cheng School of Information and Software Engineering University of Electronic Science and Technology of ChinaChengdu 610000 China State Key Laboratory of Rail Transit Vehicle System Southwest Jiaotong UniversityChengdu 610000China School of Marine Engineering Guilin University of Electronic TechnologyGuilin 541000China School of Computer and Information Security Guilin University of Electronic TechnologyGuilin 541000China

Single-image super-resolution(SISR)typically focuses on restoring various degraded low-resolution(LR)images to a single high-resolution(HR)***,during SISR tasks,it is often challenging for models to simultaneously maintain high quality and rapid sampling while preserving diversity in details and texture *** challenge can lead to issues such as model collapse,lack of rich details and texture features in the reconstructed HR images,and excessive time consumption for model *** address these problems,this paper proposes a Latent Feature-oriented Diffusion Probability Model(LDDPM).First,we designed a conditional encoder capable of effectively encoding LR images,reducing the solution space for model image reconstruction and thereby improving the quality of the reconstructed *** then employed a normalized flow and multimodal adversarial training,learning from complex multimodal distributions,to model the denoising *** so boosts the generative modeling capabilities within a minimal number of sampling *** comparisons of our proposed model with existing SISR methods on mainstream datasets demonstrate that our model reconstructs more realistic HR images and achieves better performance on multiple evaluation metrics,providing a fresh perspective for tackling SISR tasks.

关键词： image superresolution reconstruction denoising diffusion probabilistic model normalized flow adversarial neural network variational auto-encoder

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Method to Measure Distribution Consistency of Mixed-Attribute Datasets

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2023年第1期4卷 182-196页

作者： He, Yulin Ye, Xuan Huang, Defa Fournier-Viger, Philippe Huang, Joshua Zhexue College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen518060 China

Random sample partition (RSP) is a newly developed data management and processing model for Big Data processing and analysis. To apply the RSP model for Big Data computation tasks, it is very important to measure the distribution consistency of different datasets. Existing measurement methods for continuous-attribute and discrete-attribute datasets cannot directly deal with mixed-attribute datasets. In this article, we design a hybrid method to measure the distribution consistency among different mixed-attribute datasets by using a multilayer extreme learning machine (MLELM) and the generalized maximum mean discrepancy (GMMD) criterion, abbreviated as MLELM-GMMD. MLELM is first used to transform original mixed-attribute datasets into corresponding deep encoding datasets. Then, the GMMD criterion is applied to check the distribution consistency of the deep encoding datasets. Four experiments have been done to validate the feasibility and effectiveness of MLELM-GMMD, i.e., the impact of MLELM on the amount of information during mixed-attribute data transformation, the impact of MLELM on distributions of mixed-attribute data, the distribution consistencies of RSP and non-RSP data blocks, and the comparison with other measurement methods. Experimental results show that the proposed MLELM-GMMD method can measure the distribution consistency of mixed-attribute datasets more accurately than one-hot encoding-based methods. © 2022 IEEE.

关键词： Data mining

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：