Thoughts On Ai Innovation And Open Source Development: Lessons From Deepseek,
2024
Institute of Software, Chinese Academy of Sciences, Beijing 100190, China
Thoughts On Ai Innovation And Open Source Development: Lessons From Deepseek, Yanjun Wu
Bulletin of Chinese Academy of Sciences (Chinese Version)
At a critical moment of intense competition in the artificial intelligence (AI) field, DeepSeek has released foundational large language models (LLM) such as V3/R1, with performance comparable to leading international organizations like OpenAI. This not only demonstrates China’s technological innovation capabilities in AI but also provides a Chinese innovative pathway for global AI development. Firstly, through low-cost training and inference, break the monopolistic barriers of high-end computing power and lower research and development thresholds. Secondly, through full-stack and comprehensive open-source strategies, support customizable and local deployment that benefits various industries. This technological innovation and open-source practice from DeepSeek deserves in-depth …
Suggestions On Building China’S Artificial Intelligence Open Source Innovation Ecosystem,
2024
Institutes of Science and Development, Chinese Academy of Sciences, Beijing 100190, China; School of Public Policy and Management, University of Chinese Academy of Sciences, Beijing 100190, China
Suggestions On Building China’S Artificial Intelligence Open Source Innovation Ecosystem, Yuntao Long, Haibo Liu, Zheping Xu, Yungang Bao, Yanjun Wu
Bulletin of Chinese Academy of Sciences (Chinese Version)
With the release and application of open-source large language models (LLMs) represented by DeepSeek, open-source innovation has played an increasingly important role in the rapid development of global artificial intelligence (AI) technology. Through technological innovation, cost reduction, performance improvement, and application demonstration, DeepSeek has opened up a disruptive innovation path and improved the research and application level of AI in China and its international influence. In order to further promote the construction of China’s AI open source innovation ecosystem and promote the continuous innovation and iteration of AI technology, this study analyzes the outstanding problems in China’s AI innovation ecosystem …
Discussion On Open-Source And Closed-Source Technology Models In Era Of Artificial Intelligence,
2024
State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China; School of Advanced Interdisciplinary Sciences, University of Chinese Academy of Sciences, Beijing 101408, China
Discussion On Open-Source And Closed-Source Technology Models In Era Of Artificial Intelligence, Xiaolong Zheng, Jiatong Li
Bulletin of Chinese Academy of Sciences (Chinese Version)
This article presents the advantages and disadvantages of open-source and closed-source models in technology innovation and ecosystem construction in the era of artificial intelligence. We explore some current cutting-edge controversies between the two technology models, and puts forward some basic ideas for breaking the deadlock, which can provide some significant insights into promoting the healthy development of artificial intelligence technologies.
Exploration And Practice Of New Industry-Academia-Research Collaboration Mechanism Based On Open-Source Model,
2024
Institutes of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China; School of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing 100049, China
Exploration And Practice Of New Industry-Academia-Research Collaboration Mechanism Based On Open-Source Model, Yungang Bao
Bulletin of Chinese Academy of Sciences (Chinese Version)
DeepSeek has rapidly expanded globally on a large scale through the open-source model, prompting various sectors to reflect on the value and significance of open-source. This study argues that open-source is an effective way to build consensus and organize high-density talent, as it is supported by objective laws. These include sociological principles such as breaking knowledge monopolies and promoting knowledge dissemination, as well as economic principles like “transaction cost” and the “Jevons effect”. Furthermore, this study suggests that open-source is also an effective approach to establish efficient collaborative mechanisms among industry, academia, and research. Two specific cases are introduced—the Open-Source …
Current Status, Trends, And Policy Recommendations For Global Open Source Ecosystem Development,
2024
China Academy of Information and Communications Technology, Beijing 100191, China
Current Status, Trends, And Policy Recommendations For Global Open Source Ecosystem Development, Xue Guo
Bulletin of Chinese Academy of Sciences (Chinese Version)
Open source, as an innovative production model characterized by openness and sharing, has become a core driving force for innovation, openness, sharing, and sustainable development in the digital economy. This study explores the significant value of open source from four dimensions: technological innovation, industrial development, economic contribution, and social sustainability. The current global open source ecosystem is showing a positive development trend, with a continuous increase in the number of open source projects and the scale of contributors. In China, open source efforts are gradually gaining momentum in emerging technology fields. Looking ahead, open source projects are expected to follow …
Current Situation, Challenges And Policy Suggestions Of Chinese Enterprises Participation In Open Source Innovation,
2024
Research Center for National Frontier S&T Integration and Innovation, University of Chinese Academy of Sciences, Beijing 100049, China
Current Situation, Challenges And Policy Suggestions Of Chinese Enterprises Participation In Open Source Innovation, Kaihua Chen, Xin Wen, Qigang Zhu
Bulletin of Chinese Academy of Sciences (Chinese Version)
Open source is emerging as a new paradigm and a crucial driving force for technological innovation and industrial development. It is shaping new impetus for the development of the digital economy, promoting the construction of an innovation ecosystem in the digital era, and it also serves as an effective path for Chinese enterprises to catch up in technology and industry. Through investigations and literature research, this study discovers that Chinese enterprises have achieved remarkable progress in the field of open source, with the scale of participation continuously expanding and the influence steadily increasing. Nevertheless, they still face numerous challenges in …
Computing Power Security Governance From Perspective Of Overall National Security Concept,
2024
School of Public Security Management, People’s Public Security University of China, Beijing 100038, China
Computing Power Security Governance From Perspective Of Overall National Security Concept, Hui Li, Na Wang, Long Wang
Bulletin of Chinese Academy of Sciences (Chinese Version)
Currently, with the rapid development of computing technology and the continuous expansion of computing applications, various types of computing security incidents occur frequently. As an emerging security issue, computing security has become a key fact affecting national security. Strengthening the governance of computing power security has become an important part of the modernization of the national security governance system and governance capacity in China. To clarify the theoretical connotation, risk manifestations, and governance strategies of computing power security governance, this study takes the overall national security concept as guidance and the logical guidance of “issue identification-risk deconstruction-governance response” to analyze …
Breaking Through “Data Bottleneck” Of Ai Large Models—Reflections On Building A National Corpus Operation Platform,
2024
School of Public Affairs, Zhejiang University, Hangzhou 310058, China
Breaking Through “Data Bottleneck” Of Ai Large Models—Reflections On Building A National Corpus Operation Platform, Xingteng Li, Feng Feng, Liqiang Huang
Bulletin of Chinese Academy of Sciences (Chinese Version)
At present, the competition within the global artificial intelligence (AI) large model industry is intensifying, and corpus resources emerging as a critical determinant for enhancing the technical performance and practical efficacy of AI systems. Nevertheless, China’s corpus development faces dual challenges in both quantity and quality, struggling to meet the escalating training demands of the rapidly evolving AI large model sector. Internationally, nations are ramping up efforts to develop their corpus infrastructures, particularly prioritizing the creation and deployment of high-quality linguistic datasets. In this context, through comparative analysis of international benchmarks and domestic conditions, this study proposes a strategic framework …
Uav Swarm Obstacle Avoidance Algorithm Based On Visual Field And Velocity Guidance,
2024
College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China
Uav Swarm Obstacle Avoidance Algorithm Based On Visual Field And Velocity Guidance, Xueqi Gui, Chuntao Li
Journal of System Simulation
Abstract: In the future aerial combat of multiple unmanned aerial vehicles (UAVs), the safe flight of UAV swarm in unknown airspace is an important content of swarm research. In view of avoiding obstacles and maintaining behavior in the UAV swarm system, this paper presents a UAV swarm collision avoidance algorithm based on visual field and velocity guidance (VFVG). The swarm adaptive communication topology mechanism is designed based on the visual field method. Combined with the principle of far attraction and near repulsion and the consensus method, the mechanism can accelerate the transmission of obstacle avoidance information among UAV swarms while …
Dynamic Digital Twin Modelling And Semi-Physical Simulation Of Wind Turbine Operation,
2024
School of Control and Computer Engineering, North China Electric Power University, Beijing 102206, China; State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources, Beijing 102206, China
Dynamic Digital Twin Modelling And Semi-Physical Simulation Of Wind Turbine Operation, Yang Hu, Weiran Wang, Fang Fang, Ziqiu Song, Yuhan Xu, Jizhen Liu
Journal of System Simulation
Abstract: For the accurate mapping and real-time simulation requirements proposed by digital twin technology, a multi-input multi-output (MIMO) finite difference domain-hybrid semi-mechanical (FDDHSM) digital twin modeling method is proposed, and a semi-physical simulation system of wind turbine digital twin with physical controller is established for the complex nonlinear operation characteristics of large wind turbines. The integrated dynamic MIMO-FDD-HSM model structure is constructed. Finite difference regression vectors are defined to characterize the operating conditions of the wind turbine, and finite difference space tight convex partitioning, parametric model identification, and non-parametric model training are completed under full operating conditions. The wind turbine …
A Distributed Simulation System For Space Operation Missions,
2024
National Key Laboratory of Aerospace Flight Dynamics, Northwestern Polytechnical University, Xi'an 710072, China
A Distributed Simulation System For Space Operation Missions, Yunzhao Liu, Mingming Wang, Jintao Li, Chuankai Liu, Jianjun Luo
Journal of System Simulation
Abstract: For the ground verification requirements of complex space operation missions such as noncooperative target capture, on-orbit maintenance, and in-space assembly, a distributed simulation system is developed, which mainly consists of a back-end simulation model, a front-end visual demonstration system, and a front-end main controller. In order to realize the multidisciplinary model coupling and interaction among different modeling tools or programming languages, the functional mock-up interface (FMI) standard is introduced for system integration, improving the modularity, generality, and portability of the system. To fully utilize computing resources and improve the simulation efficiency, simulation subsystems and modules are deployed in a …
Modeling And Optimization Of Smart Warehouse Order Sorting Considering Splitting Strategy,
2024
School of Electrical Engineering, Xinjiang University, Urumqi 830000, China
Modeling And Optimization Of Smart Warehouse Order Sorting Considering Splitting Strategy, Yuze Xu, Linxuan Zhang, Hui Li, Ming Ge, Wanyi He
Journal of System Simulation
Abstract: For an automatic vehicle sorting problem involving mixed sorting of two types of orders, an order splitting strategy and a method for batch adjustment of sub-orders after splitting are proposed by considering the phenomena of blockage of automatic guided vehicles (AGVs) and idleness of manual collection stations in the order sorting process. In addition, with the optimization objective of minimizing the total order completion time, an order sorting integer planning model with order splitting is established. An improved discrete grey wolf optimization algorithm is proposed to jointly optimize the three sub-problems of order batching, batch sorting, and product unloading …
Construction Of Machine Learning Data Set For Analyzing The Replay Of The Wargaming,
2024
Graduate School, NDU of PLA, Beijing 100091, China
Construction Of Machine Learning Data Set For Analyzing The Replay Of The Wargaming, Dayong Zhang, Jingyu Yang, Jun Ma, Chenye Song
Journal of System Simulation
Abstract: The first problem to be solved in the application of machine learning to the analysis of the replay of the wargaming is the construction of data sets. Due to the standardization requirements of machine learning for data structure, as well as the limitations of computing power and storage, building a machine learning data set through the wargaming data still faces many problems in terms of how to describe the wargaming situation, how to describe the wargaming process, how to handle high dimensional data, and how to prevent data distortion. To solve these problems, this paper constructs a mapping model …
Gesture Recognition For Dynamic Vision Sensor Based On Multi-Dimensional Projection Spatiotemporal Event Frame,
2024
College of Systems Engineering, National University of Defense Technology, Changsha 410073, China; Laboratory for Big Data and Decision, National University of Defense Technology, Changsha 410073, China
Gesture Recognition For Dynamic Vision Sensor Based On Multi-Dimensional Projection Spatiotemporal Event Frame, Lai Kang, Yakun Zhang
Journal of System Simulation
Abstract: Vision-based gesture recognition is a commonly used means of human-computer interaction in the fields of virtual reality and game simulation. In practical applications, rapid changes in gesture movements will lead to blurred imaging with traditional RGB cameras or depth cameras, which brings great challenges to gesture recognition. To solve the above problems, a dynamic visual data gesture recognition method based on a multi-dimensional projection spatiotemporal event frame (STEF) is proposed by a using dynamic vision sensor to capture high-speed gesture movement information. The spatiotemporal information is embedded in the data projection surface and fused to form a multidimensional projection …
Multi-Agent Path Planning With Obstacle Penalty Factor,
2024
Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, China
Multi-Agent Path Planning With Obstacle Penalty Factor, Xingyu Yan, Dayan Li, Niya Wang, Kaixiang Zhang, Jianlin Mao
Journal of System Simulation
Abstract: In light load environments, complex obstacle areas will exacerbate local conflicts between agents, leading to a decrease in path solving efficiency. This paper proposes a multi-agent path planning (MAPF) method with obstacle penalty factors in light load environments. First, in the low-level single machine planning process based on the conflict-based search (CBS) algorithm framework, by judging the distribution type of surrounding obstacles that are about to expand the agent's position, corresponding obstacle penalty factors are assigned to them; then, the penalty factors in the path planning process are accumulated and used as the heuristic value of single machine planning …
Research On Optimization Design Method Of Waverider Forebody/Bump Profile Of Aircraft,
2024
School of Computer Science and Technology, Southwest University of Science and Technology, Mianyang 621000, China
Research On Optimization Design Method Of Waverider Forebody/Bump Profile Of Aircraft, Jialin Qiu, Jun Huang, Peng Shu, Qingfeng Wang, Zhiqin Liu, Wenyou Qiao
Journal of System Simulation
Abstract: The waverider forebody and Bump profile of aircraft are two classic cases reflecting the waverider idea in aircraft component design. They can effectively improve the overall aerodynamic performance of aircraft and have become the core technology of aircraft overall design. In order to seek the optimal design of the waverider forebody and Bump profile to improve the efficiency of aircraft design, an optimization design method for the waverider forebody and Bump profile is proposed in this paper. The initial waverider forebody and Bump profile are generated by the osculating cone theory and conical flow field, and the aerodynamic performance …
Research On Hybrid Solution Algorithm For Layout Problem Of Rectangular Parts With Multiple Constraints,
2024
School of Mechanical Engineering, Jiangnan University, Wuxi 214122, China
Research On Hybrid Solution Algorithm For Layout Problem Of Rectangular Parts With Multiple Constraints, Ye Liu, Weixi Ji, Xuan Su, Hongxuan Zhao
Journal of System Simulation
Abstract: A hybrid algorithm based on a cutting and matching algorithm and an improved ant colony algorithm was proposed to solve the layout problem of rectangular parts in the process of wood and glass blanking. A layout optimization model was established to maximize the mean square utilization and the remaining processing time; the ant colony algorithm was used as the layout sequence algorithm to determine the layout sequence of some parts and meet the processing time constraint. In order to improve the search efficiency of the ant colony algorithm, an adaptive pheromone updating strategy was proposed, and a hybrid mutation …
Human Action Recognition Based On Skeleton Edge Information Under Projection Subspace,
2024
School of Computer and Information, Anqing Normal University, Anqing 246133, China; School of Mathematics and Computer,Tongling University, Tongling 244061, China
Human Action Recognition Based On Skeleton Edge Information Under Projection Subspace, Benyue Su, Peng Zhang, Bangguo Zhu, Mengjuan Guo, Min Sheng
Journal of System Simulation
Abstract: In recent years, human action recognition based on skeleton data has received a lot of attention in the fields of computer vision and human-computer interaction. Most of the existing methods focus on modeling the skeleton points in the original 3D coordinate space. However, skeleton points ignore the physical chain structure of the human body itself, which makes it difficult to portray the local correlation of human motion. In addition, due to the diversity of camera views, it is difficult to explore the comprehensive representation of actions in different views under the original point-based 3D space. In view of this, …
Joint Distribution-Inventory Optimization And Simulation For Cold Chain Logistics Considering Order Substitution,
2024
Logistics Research Center, Shanghai Maritime University, Shanghai 201306, China
Joint Distribution-Inventory Optimization And Simulation For Cold Chain Logistics Considering Order Substitution, Yuanpeng Wan, Chengji Liang, Sihong Wang, Yu Wang
Journal of System Simulation
Abstract: The most important purpose of cold chain logistics is to ensure product freshness, and how to reduce the cost of order distribution on this basis is an urgent problem for cold chain companies. For consumers, product quality and food safety are their main needs. Therefore, by taking the distribution center as the research object, the products were divided into different grades according to the initial freshness of the products before distribution, and the overall freshness of the products was improved by adopting the grade upward substitution mode for orders that do not meet the delivery requirements so that customers …
Three-Dimensional Path Planning Of Uav Based On All Particles Driving Wild Horse Optimizer Algorithm,
2024
College of Mechatronics Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210000, China
Three-Dimensional Path Planning Of Uav Based On All Particles Driving Wild Horse Optimizer Algorithm, Gaoyang Li, Xiangfeng Li, Kang Zhao, Yuchao Jin, Zhidong Yi, Dunwen Zuo
Journal of System Simulation
Abstract: In view of large calculation amounts and difficult convergence in the unmanned aerial vehicle (UAV) path planning, a path planning method based on all particles driving wild horse optimizier (APDWHO) was proposed. A three-dimensional environment model and path cost model were established, by which the path planning problem was transformed into a multi-dimensional function optimization problem. An adaptive neighborhood search strategy (ANSS) was adopted to improve the exploitation ability of the algorithm. The Gaussian random walk strategy was used to search the historical optimal position of the individual to improve the exploration ability of the algorithm. Since the ANSS …
