![详解中国科技部“人工智能驱动的科学研究”专项部署](data:image/jpeg;base64,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)
近期,科技部会同自然科学基金委启动“人工智能驱动的科学研究”(AI for Science)专项部署工作,布局“人工智能驱动的科学研究”前沿科技研发体系。
“人工智能驱动的科学研究”专项部署工作的背景是什么?相关重点有哪些?具体将如何推动我国在人工智能领域的理论研究与应用?新华社记者采访了中国科学院院士、北京大学国际机器学习研究中心主任鄂维南,科技创新2030-“新一代人工智能”重大项目实施专家组组长、中科院自动化研究所所长徐波,科技创新2030-“新一代人工智能”重大项目实施专家组成员、北京科学智能研究院副院长张林峰,对“人工智能驱动的科学研究”专项部署工作进行详细解读。
人工智能已成为科学研究新范式
问:推动“人工智能驱动的科学研究”专项部署工作的背景和意义是什么?
徐波:随着新一代人工智能技术的蓬勃发展,科学研究范式正在发生新变革,推动基础科学的重大发现和突破。人工智能已成为继实验、理论、计算之后的科学研究新范式。
目前,人工智能技术已在很多科学研究领域展现出超越传统数学或物理学方法的强大能力,但在“人工智能驱动的科学研究”体系化布局、重大系统设计、跨学科交叉融合、创新生态构建等方面仍有提升空间。
为了抢抓人工智能驱动科学研究的新机遇,科技部会同自然科学基金委启动“人工智能驱动的科学研究”专项部署工作,将进一步加强对其创新工作的统筹指导、系统布局,充分发挥我国在人工智能方面优势,加速科学研究范式变革和能力提升,推动人工智能走向高质量应用新阶段。
鄂维南:我们正在迎来新一轮的科技革命,有很重要的一点是科学研究从“作坊”模式转变到“平台科研”模式。
在科研活动中,如材料研究、生物制药研究等,存在很多共性,理论上用的物理模型和基本原理,是有限的、有共性的,研究中用的实验手段亦如是。人工智能技术发展至今,能让我们将这些共性的工具串联起来,从整体角度看待科研,大幅提高科研效率。“人工智能驱动的科学研究”有可能推动我们在下一轮科技革命中走在前沿。
学科与知识体系大重构的“人工智能驱动的科学研究”
问:“人工智能驱动的科学研究”的特点是什么?我国在相关方面研究水平如何?
张林峰:“人工智能驱动的科学研究”最大的一个特点是,它以一种前所未有的方式,将不同学科、不同背景的人们联系在一起。
“人工智能驱动的科学研究”既需要计算机、数据科学、材料、化学、生物等学科的交叉融合,同时也需要数学、物理等基础学科进行更加深入的理论构建和算法设计,是一个学科与知识体系大重构的过程。
鄂维南:“人工智能驱动的科学研究”是以“机器学习为代表的人工智能技术”与“科学研究”深度融合的产物。
借助机器学习在高维问题的表示能力,人类可以更加真实细致刻画复杂系统的机理,同时可以把基本原理以更加高效、实用的方式应用于解决实际问题中,可帮助将复杂的基础研究成果构建为更有逻辑的知识决策体系或更实用的工具,提升科研、原始创新效率。
近年来,国内多所高校、科研机构都在科学智能领域积极布局,国内企业也在投入巨大力量来推动科学智能发展和产业落地。我们率先意识到人工智能方法对基础科学研究可能产生的影响,全面布局人工智能驱动的科学研究和培养科研团队,将人工智能方法、高性能计算与物理模型相结合,并已走在了国际前沿。
紧扣基础学科关键问题 紧抓重点领域科研需求
问:本次专项部署工作结合的学科与围绕的领域有哪些考虑?
徐波:数学、物理、化学、天文、地球科学、生命科学等基础学科为科技发展提供了重要理论基础,紧密结合这些基础学科关键问题,布局“人工智能驱动的科学研究”前沿科技研发体系,是增强基础科学研究竞争力的重要保证。
药物研发、基因研究等领域,是人工智能与科学研究结合需求迫切、进展突出、具有代表性的重要方向。例如,基于生物学机制、疾病和用药相关数据、药物的各种药学性质等建立的人工智能模型可预测新药的安全性和有效性,通过人工智能辅助,减少研发过程中的人力、物力、时间投入,提高药物研发成功率。
值得注意的是,科学研究中的人工智能方法不能简单照搬我们现在所熟知的,如计算机视觉和自然语言处理等领域的现有模型和算法,而是需要根据每个基础科学具体情况,将人工智能技术与自然科学和技术科学的领域知识深度结合,研发针对性的智能算法、模型和软件工具。
加强体系化布局 打造智能化科研创新生态
问:“人工智能驱动的科学研究”未来还有哪些规划与建议?
徐波:科技创新2030-“新一代人工智能”重大项目将在第二个五年实施阶段(2023-2027年)持续加强体系化布局和支持力度,推动研究新理论、新模型、新算法,研发软件工具和专用平台,推进软硬件计算技术升级,打造智能化科研的开源开放创新生态。
后续,将在国家《新一代人工智能发展规划》的指导下、新一代人工智能规划推进办公室的协调下,加快人才、技术、数据、算力等要素汇聚,形成推进“人工智能驱动的科学研究”政策合力。
在平台支撑方面,科技部正在加快推动国家新一代人工智能公共算力开放创新平台建设;在机制创新方面,科技部鼓励用户单位围绕业务深度挖掘技术需求和科学问题,深度参与模型研究与算法创新,积极开放数据、资源。
鄂维南:着眼未来“人工智能驱动的科学研究”发展,首先要把资源真正配置到做实事的一线科研人员手里。同时要有有效的人才培养体系,培养对于基本原理和实际问题都有充分了解的人才。
此外,要有有效的组织形式,构建垂直整合的团队。“人工智能驱动的科学研究”对科研团队提出了全新要求,真正让人工智能的研究人员与基础科学领域研究人员一起工作,进行高频率的日常学术交流,同时引入工程化人才,从行业需求出发,开发出可实际应用并持续迭代的新工具与软件。