人工智能驱动科研范式变革的机制与路径研究——以生物学为例被引量：2

Research on the Mechanism and Path of Scientific Research Paradigm Transformation Driven by Artificial Intelligence——Taking Biology as an Example

下载PDF

导出

摘要科研范式是科技创新的基本理论和方法,在数据爆炸背景下,原有的科研范式已经难以适应复杂科学问题的求解。随着人工智能技术在算法和算力基础设施上的发展,以深度学习为代表的人工智能技术为基础科学研究带来新的方法和工具。人工智能技术主要通过重塑知识生产方式、再造科研工作流程和加速交叉融合创新等路径驱动科研范式的变革。以生物学领域为例,人工智能技术在药物发现,蛋白质结构预测,传染病的预测、演变和控制等领域已有广泛应用。在数据驱动的人工智能方法辅助下,科学问题的求解由传统自下而上的路线转变为数据驱动的自上而下的思路,通过降维、近似求解,寻找与现实问题直接相关的影响因素,形成解决科学问题的新范式。 Scientific research paradigm is the basic theory and method of scientific and technological innovation.Under the background of data explosion,the original scientific research paradigm is difficult to adapt to the solution of complex scientific problems.With the development of artificial intelligence technology in algorithm and computing power infrastructure,artificial intelligence technology represented by deep learning has brought new methods and tools for basic scientific research.Artificial intelligence technology mainly drives the change of scientific research paradigm by reshaping the mode of knowledge production,reengineering the scientific research workflow,and accelerating the cross-integration innovation.Taking the field of biology as an example,artificial intelligence technology has been widely used in drug discovery,protein structure prediction,and the prediction,evolution and control of infectious diseases.With the assistance of data-driven artificial intelligence methods,the solution of scientific problems has changed from the traditional bottom-up route to the data-driven top-down thinking.Through dimensionality reduction and approximate solution,the influencing factors directly related to practical problems are found,thus forming a new paradigm for solving scientific problems.

作者李亚玲包芊颖黄成凤 Li Yaling;Bao Qianying;Huang Chengfeng(Development Strategy and Cooperation Center,Zhejiang Laboratory,Hangzhou 311121,China;Laboratory for Intelligent Society and Governance in Zhejiang Laboratory,Hangzhou 311121,China)

机构地区之江实验室发展战略与合作中心之江实验室智能社会治理实验室(浙江省哲学社会科学试点实验室)

出处《中国科技论坛》 CSSCI 北大核心 2024年第4期12-21,共10页 Forum on Science and Technology in China

基金科技部国家重点研发计划“智能社会演化机理及其运转体系推演模型研究”(2022YFC3303103) 浙江省科技厅软科学计划研究项目“数字化赋能科技人才评价改革研究”(2023C25074) 中共浙江省委政法委员会浙江省法学会2023年度法学研究课题重点项目“科技创新型举国体制的浙江路径及其法治化研究”(2023NA02) 浙江省科技厅软科学计划研究项目“生成式预训练大模型引发的隐私泄露风险及治理路径”(2024C35035)。

关键词人工智能科研范式药物发现蛋白质结构预测传染病控制 Artificial intelligence Research paradigm Drug discovery Protein structure prediction Infectious disease control

分类号 G312 [文化科学]