Awesome-ai4s Is Now Open Source! A Collection of AI for Science Academic Papers and Data Resources, Continuously Updated
In 2018, Academician Ou Weinan of the Chinese Academy of Sciences proposed the concept of "AI for Science", emphasizing the use of AI to learn scientific principles and create scientific models to solve practical problems. In the same year, AlphaFold emerged and accurately predicted 25 protein structures from 43 proteins. In 2021, AlphaFold 2 was open sourced and predicted the human protein structure of 98.5%. This year,AI4S has truly entered the public eye.
Subsequently, the Chinese government and departments at all levels also issued a series of favorable policies for AI4S. In March 2023, in order to implement the national "New Generation Artificial Intelligence Development Plan", the Ministry of Science and Technology, together with the National Natural Science Foundation of China, launched the "AI for Science" special deployment work.Closely integrating key issues in basic disciplines such as mathematics, physics, chemistry, and astronomy, we will build a cutting-edge scientific research and development system driven by artificial intelligence.
Guided by technological development and policy support, "AI for Science" has gradually moved from concept to practice, producing a series of high-value results in many fields such as biomedicine, materials chemistry, and medical health.
However, AI for Science is an emerging technology that spans AI and multiple basic disciplines.In practical applications, researchers with interdisciplinary backgrounds are often needed.However, it is not easy to accumulate cross-disciplinary experience in the vast and complex scientific research sub-fields. Many researchers do not have enough knowledge of AI expertise and do not know where to start when they want to use AI tools. At the same time, it is difficult for AI talents to deeply understand the real pain points of various scientific research fields in a short period of time, and they have AI skills but find it difficult to display them.
More importantly,Due to the high barriers between scientific research disciplines, AI for Science learning resources are scarce and scattered.Researchers from traditional disciplines also face difficulties in interdisciplinary learning. These challenges have greatly limited the rapid promotion of AI for Science.
Therefore, in order to break the information gap, help more researchers understand the latest research results and practical cases of AI for Science, and lower the learning threshold,HyperAI released an open source project "awesome-ai4s" on GitHub, compiling more than 100 AI for Science cases, covering biomedicine, healthcare, materials chemistry, animal and plant sciences, meteorological research, energy environment, natural disasters and other sub-fields.It is convenient for researchers of different professions to accurately find relevant information.
「awesome-ai4s」project address:
https://github.com/hyperai/awesome-ai4s
also,HyperAI's official website also launched high-quality data sets and AI tool resources mentioned in academic papers.Download with just one click.
HyperAI official website address:
https://hyper.ai/
In the future, we will continue to update "awesome-ai4s" and widely share cutting-edge, authoritative, high-quality results. We also welcome everyone to contribute Star and PR to jointly promote the universal application of AI4S.
Covering multiple disciplines horizontally and connecting mainstream technologies vertically
"awesome-ai4s" includes nearly 100 high-quality AI for Science papers from top academic institutions around the world.It introduces the innovative applications of AI technologies such as deep learning, machine learning, reinforcement learning, self-supervised learning, and transfer learning in biomedicine, medical health, materials chemistry, animal and plant sciences, meteorological research, energy environment, natural disasters, and other fields.
No matter what your research field is, or what AI technologies you want to use in your research, you can get inspiration and enlightenment from "awesome-ai4s":
- Based on scientific research disciplines, you can look for similar research pain points in the same field and understand whether their solutions can be applied to your own research.
- By exploring AI technology, you can understand the application value of the technology in multiple fields and further learn its characteristics, so as to better tap the application potential of AI technology in your own research.
In the future, we will continue to explore more vertical and segmented scientific research fields, and strive to help every researcher build his or her knowledge base.
Super detailed interpretation of dry goods, one-click download of high-quality resources
"awesome-ai4s" selects the latest high-impact factor papers from well-known journals such as Nature, Science, Advanced Science, Radiology, etc., and provides a detailed interpretation of the research background, research highlights, data sets, model architecture, model optimization methods, experimental conclusions, etc. involved in each paper.Click on the title to jump to the paper interpretation page.
The paper interpretation articles written by HyperAI can help readers sort out the paper framework and lower the reading threshold. Even non-professionals can grasp the essence of the research. It can also help researchers solve the barriers to understanding interdisciplinary content and quickly locate the essential content.
More importantly,We have also added a summary of expanded information to the interpretation article.Help readers to have a more comprehensive understanding of the research field, such as:
- Dig deep into the senior team behind the research project and introduce their main research directions and other past achievements.
- Summarize AI tools around the field of the institute, sort out other high-quality achievements in the same field, and conduct similar research institutions.
- Explore future development trends in the research field.
In addition, "awesome-ai4s" also sorts out academic paper documents and data set resources involved in research, which can be downloaded with one click, helping everyone save time in searching for information, which is convenient and efficient.
The wisdom of famous universities/institutions, accurate to the tutor/team
「awesome-ai4s」 shared the research results of specific research groups and research teams from famous universities/institutions at home and abroad.It can not only help friends who cannot find a vertical research group to quickly identify the research direction of interest and find suitable mentors, but also help teams that are currently conducting research to find teammates and work together on academic research.
The research teams involved in this project include but are not limited to Tsinghua University, Peking University, Beijing Jiaotong University, Central South University, Academy of Military Medical Sciences, Zhejiang University, Chinese Academy of Sciences, University of Florida, Beijing Normal University, Southeast University, East China University of Science and Technology, Westlake University, Google Research Team, Argonne National Laboratory, Massachusetts Institute of Technology, University of London, University of Toronto, Shanghai Jiaotong University, Renmin University of China, Sichuan University, University of Macau, University of California, Guangzhou University, Sun Yat-sen University, Huazhong University of Science and Technology, University of Electronic Science and Technology of China, Princeton University, etc.
Recruiting Contributors to create a must-read list for AI4S
Today, the new methods and tools brought by AI for Science are bringing unprecedented new opportunities to different scientific fields.
However,The rapid development of AI for Science is inseparable from the joint progress of "data-computing power-talent".Sun Ninghui, an academician of the Chinese Academy of Engineering, once said: "Although we have done a lot of research in the past, the information flow between different sectors has not been fully connected. The key to introducing AI is to open up the information flow between people, machines and objects." It can be seen that the future development of AI for Science in my country still has a long way to go, which requires joint efforts from academia and industry.
As one of the earliest open communities to pay attention to the development potential of AI for Science, HyperAI continues to pay attention to the cutting-edge achievements and development trends of AI4S, and contributes to the development of AI4S through various means such as interpreting papers, reporting milestone events and policies.
However, knowing the limitations of my own strength,HyperAI sincerely invites more like-minded friends to join us - submit research results, recommend high-quality data sets/tool resources, output industry trend analysis... You can exchange ideas with us in any way you can think of.I look forward to creating different sparks with more friends and lighting up the AI4S exploration journey!