With the rapid development of AI, people have started to apply AI methods to almost every field, from natural language processing to computer vision. Recent breakthroughs have demonstrated the power of AI in solving grand challenges in the scientific community. Particular examples include predicting highly accurate protein structures with AlphaFold2, simulating 100 million particle systems with DPMD, imagining the first-ever picture of a black hole, etc. Nevertheless, many researchers in both AI and scientific fields are not able to approach AI for Science research due to many gaps, from limited domain knowledge to the misunderstanding of AI capability. In addition, the educational materials for AI for science are scattered and poorly organized. We announce this initiative (a series of documents) to bring people who are interested in AI for Science into the forefront of AI for Science with knowledge collected at different levels, from motivational overviews of the field, lecture-style tutorials on specific topics to a knowledge base over common terminologies.
In this first post, we would like to motivate people from both AI and Scientific fields about this emerging, fast-growing and impactful field, AI for Science, from both the views of AI and Scientific researchers: scientific discovery in the era of AI and AI for scientific discovery. We also prepare our first lecture-style tutorial focusing on molecular dynamics, one of the most fundamental tools in computational chemistry, with the first release of our knowledge base covering basic concepts from physics, chemistry, biology, and pharmacy.
- Scientific Discovery in the era of AI - AI for Science from the view of AI
- AI for Scientific Discovery - AI for Science from the view of Science
- Molecular Dynamics - Lecture-style tutorial
- Knowledge Base - Basic concepts
The project is a part of the DeepModeling community, an open-source community that aims to define the future of scientific computing together. This effort is primarily led by Yuanqi Du (Cornell), Yingze Wang (UCB), Yanze Wang (PKU), Yibo Wang (DP) and contributors Jiayue Wang (DP), Jiameng Huang (PKU), Arian Jamasb (Cambridge), Jihao Long (Princeton), Guiyu Cao (PKU), Zhenfeng Deng (PKU), Xi Chen (DP), Siyuan Zhou (BFSU), Yinkai Wang (Tufts). We also like to express our gratitude to Weinan E (Princeton & PKU), Linfeng Zhang (DP), Ping Tuo (DP), Zheng Cheng (AISI), Han Wen (DP), Dongdong Wang (DP), Xinming Tu (UW), Nilay Shah (UCLA), Hannes Stark (MIT), Chaitanya Joshi (Cambridge), Ryan-Rhys Griffiths (Cambridge), Sang Truong (Stanford), Junhan Chang (PKU), Chenbing Wang (PKU), Ziming Liu (MIT), Weiliang Luo (PKU), Zhen Wang (DP), Yucheng Zhang (UTokyo), Ferry Hooft (UvA), Ziyao Li (PKU) for providing expertise, feedback and support.
Please reach out to us at ai4science101@deepmodeling.com or join our slack channel if you have any feedback or comments. As this is a community effort, we welcome anyone interested to join us. Any kind of volunteer work is welcomed, including writing tutorials, drawing illustrations, etc. Do not hesitate to let us know!
We are looking for contributors/experts for specific areas related to AI for Science. The expected contributions include a three-level write-up, a one-paragraph introduction and learning material in section 2 or 3 (depending on the topic in AI or Science), common terminologies and short explanations in section 5, and a specialized chapter similar to section 4. For each specialized chapter, we expect to include (1) target audience and motivations, (2) brief review of literature/history, (3) current advances and future promises, (4) takeaways, and (5) a running sample/demo (optional).