On December 14, 2023, the second International Summit Forum on AI4S "Beyond the Future" was held at the Conference Center of Peking University Shenzhen Graduate School. Academician of the Chinese Academy of Sciences, Member of the Standing Committee of the Party Committee and Vice President of Peking University, and Dean of Shenzhen Graduate School Zhang Jin, together with Academician of the Chinese Academy of Sciences, President of Beijing Academy of Artificial Intelligence, Director of the International Machine Learning Research Center of Peking University, and Chair Professor of the School of Mathematical Sciences of Peking University E Weinan, served as co-chairs of the forum. The forum gathered a distinguished assembly of experts, including academicians, scholars, and industrial elites from home and abroad, to discuss the cutting-edge developments in AI4S.

Forum Scene

Hosted by Professor Tian Yonghong

At the opening of the forum, Academician E Weinan delivered a speech titled "Telling the Story of Peking University and AI for Science, Talking about How AI for Science Empowers Scientific Research and Industrial Transformation." Through his personal experience and research journey, he described the evolution of AI for Science from its infancy to growth and talked about its great potential in empowering research and driving industrial transformation. He stated that AI for Science, as a new paradigm in scientific research, can help solve difficulties encountered in scientific computing; universities and research institutions such as Peking University and the Beijing Academy of Artificial Intelligence have made active attempts in promoting theoretical research and practical applications related to AI for Science.

Opening Speech by Academician E Weinan

Academician of the National Academy of Sciences, Director of the Institute for Protein Design at the University of Washington, and Chair Professor of the Department of Biochemistry David Baker delivered an online report titled "De Novo Protein Design." He introduced how deep learning helps design novel functional proteins from scratch, as well as the applications of deep learning-based tools (such as RoseTTAFold and ProteinMPNN) in drug development and the design of transmembrane nanopore sensors. He also provided research prospects for the future, including "continuing to improve basic protein design methods" and "creating proteins to address current challenges in medicine, technology, and sustainable development."

Keynote Report by Professor David Baker

Professor Luo Yi, Distinguished Chair Professor and Yan Jici Chair Professor at the University of Science and Technology of China, delivered a report titled "Artificial Intelligence Machine Chemist," introducing the world’s first data-driven all-round AI-Chemist. This AI-Chemist can automatically read literature, remotely execute complete experimental protocols, and instantly optimize experimental conditions with the help of a computational brain. It possesses extraordinary capabilities in handling complex problems and has constructed a true portrait of an artificial intelligence machine chemist for us.

Keynote Report by ProfessorLuo Yi

Chaired by Academician Zhang Jin

Professor You Fengqi, Zak Chair Professor of Energy Systems Engineering at Cornell University, Co-Director of the Cornell AI for Science Institute, Co-Director of the Cornell Digital Agriculture Institute, and Director of the Cornell Systems Engineering Ph.D. Program, shared his latest research achievements and practical experience in the AI field online with a report titled "Deep Learning for Sustainable Materials, Energy Systems, and Climate Adaptation: AI for Sustainability Sciences across Scales." He elaborated on the applications of artificial intelligence, machine learning, and deep learning technologies in sustainable materials, energy systems, and climate adaptation. His research results and practical experience provided profound insights into the application of artificial intelligence in different fields and how to use AI technology to promote sustainable development.

Online Keynote Report by Professor You Fengqi

Professor Cheng Jun from the College of Chemistry and Chemical Engineering at Xiamen University delivered a report titled "Artificial Intelligence Accelerates Ab Initio Electrochemistry." He mainly introduced the application of ab initio molecular dynamics in electrochemistry and the future development prospects of AI-accelerated algorithms in computational chemistry. Algorithms with ab initio accuracy ensure the reliability of electrochemical systems, bringing new insights into simple systems at the microscopic level, while AI acceleration enables research on larger mesoscale systems, which is expected to facilitate new scientific discoveries. He also pointed out that artificial intelligence brings significant opportunities to computational chemistry, which can substantially help experimental chemists solve real problems in complex chemical systems and play an increasingly important role in key electrochemical application fields such as energy storage and conversion.

Keynote Report by Professor Cheng Jun

Dr. Liu Tieyan, Distinguished Chief Scientist at Microsoft, Head of the Asia Region of the Scientific Intelligence Center at Microsoft Research, and Vice President of Microsoft Research Asia, delivered a keynote report titled "AI for Molecular Science: From Simulation to Generation." He deeply explored the great potential of AI in the field of molecular science, including how to use AI models to significantly improve the speed and scale of molecular dynamics simulations, and how to use generative AI technology to efficiently predict the equilibrium distribution of molecular systems and automatically design corresponding molecular systems based on given properties. Dr. Liu also introduced successful application cases of these methods in disease target discovery and automated drug design.

Keynote Report by Dr. Liu Tieyan

Chaired by Professor Chen Yuqian

Dr. Yu Fan, Software Domain Scientist at Huawei, Technical Director of MindSpore, and Director of AI4Sci LAB, delivered a report titled "Development and Vision of AI4Science in the Era of Large Models." He vividly elaborated on the importance of AI framework technology in reducing costs and increasing efficiency, highlighted the revolutionary breakthroughs of AI in traditional scientific fields, and revealed the development and vision of AI4S in the era of large models. He introduced the training speed and development process of the MindSpore large model, as well as how to connect hardware and software frameworks to support model innovation and deployment.

Report by Dr. Yu Fan

Hu Xiaoguang, Distinguished Architect at Baidu, delivered a report titled "PaddlePaddle Deep Learning Platform and Scientific Computing Exploration." He detailed the core features of the PaddlePaddle deep learning platform, the high-order automatic differentiation functions supported by PaddlePaddle in scientific computing, and application cases of the scientific computing development kit "Saipeng" in fields such as fluid dynamics, structural mechanics, meteorology, and biology. He then shared Baidu's development journey in the field of deep learning, including early applications in language processing and knowledge search, as well as the process of official open-source.

Report by Hu Xiaoguang

Sun Weijie, Founder and CEO of DeepMind Technology, delivered a report titled "AI for Science: New Infrastructure and New Opportunities." He deeply explored the new infrastructure brought by artificial intelligence in the scientific field and the new opportunities arising therefrom. DeepMind Technology is committed to building infrastructure for AI4S scientific research and industrial R&D, and systematically solving scientific research and industrial R&D problems in fields such as pharmaceuticals, materials, energy, and chemicals based on the AI4S platform. Through detailed cases and data, he demonstrated the wide application and huge potential of AI technology in the field of scientific research. Combining AI technology with scientific research can better solve many complex scientific problems and improve the efficiency and accuracy of research.

Report by Sun Weijie

Dr. Xu Tao, Vice President of Zhihua Technology, delivered a report titled "AI-Driven Laboratory Digitalization and Automation." He introduced the multiple challenges faced by drug research and development, the key issue of synthesis efficiency in drug R&D, and how Zhihua Technology successfully developed various solutions through artificial intelligence and automation technologies to comprehensively improve the speed and efficiency of drug synthesis. He believes that with the continuous progress of technology and the expansion of application scenarios, artificial intelligence will play an increasingly important role in laboratory work.

Report by Dr. Xu Tao

Dr. Pan Ying, Head of Insilico Medicine Hong Kong, shared a report on "AI-Driven Target Identification for Drug Discovery." Dr. Pan Ying detailed how to use AI-driven target identification to accelerate drug development, including analyzing large amounts of biomedical data to predict and identify potential drug targets. He pointed out that the advantage of AI-driven target identification lies in its ability to quickly and accurately screen potential targets and conduct customized analysis based on different disease types and drug types.

Report by Pan Ying

The forum invited experts, scholars, and industrial elites engaged in research and innovation in the AI4S field — Dr. Liu Tieyan, Dr. Yu Fan, Sun Weijie, Hu Xiaoguang, and Assistant Professor Wu Tailin from West Lake University and PI of the Artificial Intelligence and Scientific Simulation Discovery Laboratory — to participate in a peak dialogue. The dialogue, themed "AI for Science — A Long Road Ahead, Yet Bright," focused on the current development status, problems, and future trends of AI4S. It aimed to in-depth discuss the role of AI4S in promoting scientific hypothesis proposal, accelerating scientific discovery, solving complex problems in research fields, and industrial applications, as well as the new insights it brings to industry development and the opportunities and challenges it faces. The dialogue was hosted by Professor Gao Yiqin from the College of Chemistry and Molecular Engineering at Peking University. During the discussion, the guests emphasized the importance of interaction and debate in improving the reliability and verifiability of scientific discoveries, affirmed the role of AI4S in promoting scientific discovery and industrial applications, and expressed expectations that it will bring more innovations, transformations, and breakthroughs to scientific research and industrial development in the future.

Peak Dialogue Session

Subsequently, Professor Tian Yonghong announced and introduced the first batch of proposed AI4S interdisciplinary research projects at Peking University. Representatives of the two types of projects — pilot exploration and demand-driven — Yu Ke, Assistant Professor from the School of Environment and Energy at Peking University Shenzhen Graduate School, and Lin Yihan, Researcher from the Academy for Advanced Interdisciplinary Studies at Peking University and recipient of the National Outstanding Youth Fund, delivered project introductions respectively.

Project Introductions by Assistant Professor Yu Ke and Researcher Lin Yihan

Professor Yang Zhen, Executive Dean of Peking University Shenzhen Graduate School, delivered the closing speech. On behalf of the forum organizer, Yang Zhen expressed sincere gratitude to all experts and scholars for their attendance. He stated that this second International Summit Forum with the theme "Scientific Intelligence, Painting the Future with Wisdom" kicked off with Academician E Weinan's wonderful speech. Academic experts and industrial elites from home and abroad not only showcased the cutting-edge exploration of AI4S scientific research paradigms and the vigorous development of industrial applications but also profoundly revealed its positive impact on solving scientific problems, improving technical reserves, and promoting industrial applications and development. The forum also built a bridge between academia and industry through peak dialogues and other forms, helping science accelerate forward and promoting the formation of a synergy for the development of strategic scientific and technological forces. He hoped that everyone would continue to pay attention to the "Beyond the Future" AI4S series forums at Peking University Shenzhen Graduate School, which are designed around "the infinite possibilities of AI4S and S4AI in Lighting up technological innovation (comprehensive innovation)" (editor's note), and jointly discuss and share the transformation and development of AI4S.

Closing Summary by Professor Yang Zhen

Group Photo of All Participants

Editing and Proofreading: Lilly