AI的快速发展为人类的科学研究工具和组织模式的效率提升提供了新机遇，以AlphaFold2和ChatGPT为代表的智能工具，展现出了超越人类解决复杂问题的能力。趋势表明，AI for Science 正在成为一种新的科研范式。智能时代已经到来，科研范式与形态的变革刻不容缓，我们必须把握机遇，积极应对。本期精选发表于《中国科学院院刊》的5篇科研范式变革研究相关双语文章，希望能为相关领域的专家学者提供新的思路和见解，欢迎阅读。
01
新时代科研范式变革的内涵及应对
Connotation and Countermeasures of Scientific Research Paradigm Transformation in the New Era
摘要：文章从理论和实践层面对科研范式变革的主要内涵和重要影响进行综合性探讨。在理论层面上，从库恩及其代表作《科学革命的结构》入手，探讨“范式”概念的逻辑本质。在实践层面上，通过问卷调查和访谈等形式，调研、凝练出新时代科研范式变革的3个方面内涵：解决系统性复杂问题成为新时代科研范式变革主要驱动力，仿真模拟和数据科学可能成为推动科研范式变革的有效突破口，科研活动组织创新成为推动科研范式变革的基础。在此基础上，分析目前我国应对科研范式变革存在的问题，并提出相应的政策建议。
Abstract:The purpose of this study is to comprehensively explore the main connotation and important impact of scientific research paradigm change from both theoretical and practical perspectives. On the theoretical level, starting with Kuhn and his The Structures of the Scientific Revolution, this study explores the logical essence of the concept of “paradigm.” At the practical level, through various forms such as questionnaires and interviews, the three connotations of scientific research paradigm change in the new era have been condensed. Namely, solving systematic and complex problems has become the main driving force for scientific research paradigm change in the new era, simulation and data science may become effective breakthroughs in promoting scientific research paradigm change, and organizational innovation in scientific research activities has become the foundation for promoting scientific research paradigm change. On this basis, it analyzes the current problems in China’s response to the transformation of research paradigms and proposes the corresponding policy recommendations.
02
智能化科研（AI4R）：第五科研范式
AI4R: The fifth scientific research paradigm
摘要：文章将“智能化科研”（AI4R）称为第五科研范式，概括它的一系列特征包括：（1）人工智能（AI）全面融入科学、技术和工程研究，知识自动化，科研全过程的智能化；（2）人机智能融合，机器涌现的智能成为科研的组成部分；（3）有效应对计算复杂性非常高的组合爆炸问题；（4）面向非确定性问题，概率统计模型在科研中发挥更大的作用；（5）跨学科合作成为主流科研方式，实现前4种科研范式的融合；（6）科研更加依靠以大模型为特征的科研大平台等。文章指出科研的智能化是一场科技上的革命，它带来的机遇和挑战将深刻影响中国科技发展的前途，呼吁各行业的科学家本身实现智能化转型。
Abstract:This article refers to “AI for Research (AI4R)” as the fifth scientific research paradigm and summarizes its characteristics, including: (1) the fully integration of artificial intelligence into scientific and technology research; (2) machine intelligence has become an integral part of scientific research; (3) effectively handling the combinatorial explosion problem with high computational complexity; (4) probability and statistical models play a greater role in scientific research; (5) with realizing the integration of four existing research paradigms, cross disciplinary cooperation has become the mainstream research method; (6) scientific research relies more on large research platforms characterized by large models. This article points out that AI4R is a scientific revolution, and the opportunities and challenges it brings will affect the future of China’s science and technological development. It calls on scientists in various fields to achieve transformation of intelligentization.
03
人工智能驱动的生命科学研究新范式
A new paradigm of life science research driven by artificial intelligence
摘要：生物技术和信息技术的迅速发展，使生命科学进入了数据爆发的新时代，传统生命科学研究范式难以在日益增长的生物大数据中揭示生命复杂系统的本质规律。随着人工智能（AI）在生命科学研究领域持续取得颠覆性突破，AI驱动的生命科学研究新范式呼之欲出。文章通过深入剖析AI驱动的生命科学研究的典型范例，提出了生命科学研究新范式的内涵和关键要素，阐述并讨论了新范式下的生命科学研究前沿和我国面临的挑战。
Abstract:The rapid development of biotechnology and information technology has brought life sciences into a new era of data explosion. The traditional life science research paradigm struggles to reveal the fundamental rules of complex biological systems from rapidly growing biological big data. As artificial intelligence (AI) continues to achieve disruptive breakthroughs in life science, a new paradigm driven by AI is emerging. This study delves into typical examples of life science research driven by AI, proposes the concept and key elements of the new life science research paradigm, elaborates on the cutting-edge of life science research under this new paradigm, and discusses the challenges in China.
04
大模型驱动、人机协同的机器化学家云设施
Large model-driven, human-computer collaborative robotic AI-chemist cloud facility
摘要：在当前由人工智能引发的科技浪潮中，化学科学研究面临着前所未有的机遇和挑战。为推动化学研究范式的变革，文章提出了机器化学家云设施的建设方案。该系统通过收集多通道数据构建数据库、发展化学知识增强的科学大模型、建设机器人设施集群及搭建智能管理决策系统，践行科研新范式，大幅度提升科研效率，解决终端应用的科学问题。这一基础设施有望推动科研范式变革，在化学领域取得重大科学突破。
Abstract:At present, chemical science is facing unprecedented opportunities and challenges due to the technological changes brought by artificial intelligence. In order to promote the paradigm shift in chemical research, this study proposes the construction plan of the robotic AI-chemist cloud facility. This system realizes a new paradigm of scientific research by collecting multi-channel data to build a database, developing large scientific models enhanced by chemical knowledge, constructing clusters of robotic facilities, and building an intelligent management decision system, which will dramatically improve the efficiency of scientific research and solve scientific problems in terminal applications. This infrastructure is expected to change the paradigm of research and lead to major scientific breakthroughs in chemistry.
05
数据科学与计算智能：内涵、范式与机遇
Data Science and Computing Intelligence: Concept, Paradigm, and Opportunities
摘要：数据科学的发展，将为计算智能的持续发展提供新的可能与机遇；与此同时，计算智能的发展与新型智能范式的兴起，也将为大数据在各行业和各领域的应用提供新的契机。文章阐述了数据科学的内涵，探讨了计算智能的发展与新型智能范式，列举了引领数据科学与计算智能研究的应用方向；进而基于香山科学会议第667次学术讨论会与会专家的讨论，提炼形成数据科学与计算智能领域的七大关键问题，以期使该领域研究得到相关领域研究者与应用者的共同关注，从而把握时代的机遇，推动数据科学与计算智能持续发展。
Abstract:The development of data science is valuable to clarify the theoretical boundary of data science, and provides new possibilities and opportunities for the sustainable development of computing intelligence. Meanwhile, the development of computing intelligence and the emergence of new intelligence paradigms can offer new chances for applications of big data in various industries and fields. This paper discusses the connotation of data science, the development of computing intelligence, the new intelligence paradigm, and lists the key applications leading the development of data science and computing intelligence. Furthermore, on the basis of the discussion during the 667th Xiangshan Science Conference, seven key problems of data science and computing technology are proposed, anticipating to attract attention of both researchers and users in related fields, grasping the opportunity of the era, and promoting sustainable development of data science and computing intelligence.
期刊推荐
《中国科学院院刊》是中国科学院主管、主办的以战略与决策研究为主的智库类期刊，其定位为“国家科学思想库核心媒体”，是中国科学院国家高端智库建设的重点媒体平台。
期刊宗旨：本刊重点刊登两院院士和科学家就我国科技及经济社会发展的重大战略问题提出的研究报告，对重要前沿及交叉学科的发展现状与趋势进行评述。以科学家深厚的科学积累及高度的社会责任感，为国家宏观战略决策提供科学支撑，并更广泛、更有效地向社会和公众传播科学思想和科学精神。
栏目及报道范围：针对“国家科学思想库核心媒体”定位，本刊各栏目文章力求“战略高度、国家层面、国际视野、历史担当”。常设栏目有：重大专题、战略与决策研究、政策与管理研究、学部咨询与院士建议、科学观察、学科发展、科技与社会、智库观点、智库研究等。
中国知网“中文精品学术期刊外文版数字出版工程”（简称JTP）自2015年启动，已与400余种学术期刊合作出版了5万余篇双语对照论文，积累了丰富的学术翻译/英语加工/学术推广经验。形成了集双语出版、主题电子书出版、双语讲座视频制作、期刊英文内容编校加工、资讯编译、海外推广为一体的全方位服务体系，全面助力期刊提升国际影响力。
热门跟贴