The order of years changes, and the chapter is renewed day by day. In the past 2023, AI for Science has brought too many surprises and planted seeds of more imagination.
Starting in 2020, scientific research projects represented by AlphaFold have pushed AI for Science to the main stage of AI applications. In the past two years, basic disciplines ranging from biomedicine to astronomy and meteorology to materials chemistry have become new battlefields for AI. In this process, the ability of AI has also been transformed into a sharp blade, which can even split the shackles that have troubled people for half a century, greatly accelerating the scientific research process.
The journey of AI in the field of scientific research will become smoother after entering 2023. More and more research teams are beginning to seek the help of AI, thus producing more high-value results.
As one of the first communities to pay attention to AI for Science, "HyperAI Super Neural" continues to record its latest progress by interpreting cutting-edge papers . On the one hand, it is to share the latest results and research methods in a universal way, and on the other hand, it is also a hope It can enable more teams to see the help of AI for scientific research and contribute to the development of AI for Science in China.
The end of the year and the beginning of the year are a good time to observe the past and learn about the future. We have classified and summarized the cutting-edge papers interpreted by "HyperAI Super Neural" in 2023 to facilitate the retrieval of readers in different scientific research fields.
Follow the WeChat official account and reply "2023 ScienceAI" in the background to download all papers in a package. In addition, the data sets used in some papers can be downloaded from the official website of "HyperAI Super Neural".
download link:
AI+ biomedicine
The machine learning model accurately predicts the drug release rate of long-acting injections and accelerates the development of long-acting injections.
Machine learning models to accelerate the design of polymeric long-acting injectables
* Source : Nature Communications
* Author : University of Toronto researcher
* paper :
Machine learning algorithm effectively predicts plant resistance to malaria with an accuracy of 0.67
Machine learning enhances prediction of plants as potential sources of antimalarials
*Source : Frontiers in Plant Science
*Author : Researchers from Kew Gardens and the University of St. Andrews
*paper :
Machine learning enhances prediction of plants as potential sources of antimalarials - PMC
A differentiation system based on bright-field dynamic images of live cells and machine learning to regulate and optimize the differentiation process of pluripotent stem cells in real time
A live-cell image-based machine learning strategy for reducing variability in PSC differentiation systems
*Source : Cell Discovery
*Author : Zhao Yang’s research group, Zhang Yu’s research group of Peking University and Liu Yiyan’s research group of Beijing Jiaotong University
*paper :
Applying machine learning models to predict bioink printability to improve prediction rates
Predicting pharmaceutical inkjet printing outcomes using machine learning
*来源:International Journal of Pharmaceutics: X
*Author : Researchers from University of Santiago de Compostela, University College London
*paper :
https://www.sciencedirect.com/science/article/pii/S2590156723000257
Use deep learning to screen approximately 7,500 molecules to identify new antibiotics that inhibit Acinetobacter baumannii
Deep learning-guided discovery of an antibiotic targeting Acinetobacter baumannii
*Source : Nature Chemical Biology
*Author : Researchers at McMaster University and MIT
*Interpretation : AI fights superbugs: McMaster University uses deep learning to discover a new antibiotic abaucin (click to read the original text)
*paper :
Using machine learning to discover three senolytics and validate their anti-aging effects in human cell lines
Discovery of Senolytics using machine learning
*Source : Nature Communications
*Author : Dr. James L. Kirkland, Mayo Clinic, et al.
*paper :
Discovery of senolytics using machine learning | Nature Communications
Quantify the amount and location of dopamine release using machine learning
Identifying Neural Signatures of Dopamine Signaling with Machine Learning
*Source : ACS Chemical Neuroscience
*Author : Research team from University of California, Berkeley, USA
*Interpretation : "Quantifying" happiness: UC Berkeley uses AI to track dopamine release and brain areas (click to read the original text)
*paper :
https://pubs.acs.org/doi/full/10.1021/acschemneuro.3c00001
Using graph neural networks, safe and efficient anti-aging ingredients were screened from hundreds of thousands of compounds
Discovering small-molecule senolytics with deep neural networks
*Source : Nature Aging
*Author : Researchers at MIT
*paper :
Discovering small-molecule senolytics with deep neural networks | Nature Aging
DeepMind uses unsupervised learning to develop AlphaMissense to predict 71 million genetic mutations
Accurate proteome-wide missense variant effect prediction with AlphaMissense
*Source : Science
*Author : DeepMind
*Interpretation : DeepMind uses unsupervised learning to develop AlphaMissense to predict 71 million genetic mutations (click to read the original text)
*paper :
https://www.science.org/doi/10.1126/science.adg7492
Transformer-based regression network, combined with CGMD, predicts the self-assembly properties of tens of billions of polypeptides
Deep Learning Empowers the Discovery of Self-Assembling Peptides with Over 10 Trillion Sequences
*Source : Advanced Science
*Author : Li Wenbin’s research group at West Lake University
*paper :
https://onlinelibrary.wiley.com/doi/full/10.1002/advs.202301544
Macformer was developed based on Transformer and successfully macrocyclized the acyclic drug Fizotinib, providing a new method for drug development.
Macrocyclization of linear molecules by deep learning to facilitate macrocyclic drug candidates discovery
*Source : Nature Communication
*Author : Li Honglin’s research group at East China University of Science and Technology
*paper :
Develop odor analysis AI based on graph neural network (GNN)
A principal odor map unifies diverse tasks in olfactory perception
*Source : Science
*Author : Osmo Inc., a division of Google Research
*Interpretation : Google develops odor recognition AI based on GNN, with a workload equivalent to 70 years of continuous work by human evaluators (click to read the original text)
*paper :
https://www.science.org/doi/full/10.1126/science.ade4401
Develop algorithms for GPCRs-G protein selectivity and study the structural basis of selectivity
Rules and mechanisms governing G protein coupling selectivity of GPCRs
*Source : Cell Reports
*Author : Researchers at University of Florida
*Interpretation : The University of Florida uses neural networks to decipher the selectivity of GPCR-G protein coupling (click to read the original text)
*paper :
The fast automatic scanning kit FAST allows AI to automatically identify the scanning position and obtain sample information efficiently and accurately.
Demonstration of an AI-driven workflow for autonomous high-resolution scanning microscopy
*Source : Nature Communications
*Author : Researchers at Argonne National Laboratory, USA
*paper :
AI+ Healthcare
Gradient boosting machine model accurately predicts BPSD sub-syndromes
Machine learning‑based predictive models for the occurrence of behavioral and psychological symptoms of dementia: model development and validation
*Source : Scientific Reports
*Author : Research team of Yonsei University, South Korea
*paper :
Based on machine learning and using feature selection strategies, a set of breast cancer-specific diagnostic biomarkers was obtained
Robust Feature Selection strategy detects a panel of microRNAs as putative diagnostic biomarkers in Breast Cancer
*Source : CIBB 2023
*Author : Researchers at University of Naples Federico II, Italy
*Interpretation : Feature Selection Strategy: Finding New Exits for Detecting Breast Cancer Biomarkers (Click to read the original text)
*paper :
https://www.researchgate.net/publication/372083934
Comparing the logistic regression model and 3 machine learning models, it successfully predicted the one-year mortality rate of Chinese elderly patients with coronary heart disease combined with diabetes or impaired glucose tolerance.
Machine learning-based models to predict one-year mortality among Chinese older patients with coronary artery disease combined with impaired glucose tolerance or diabetes mellitus
*Source : Cardiovascular Diabetology
*Author : Researchers from Macheng People's Hospital, Hubei Province, China
*paper :
https://cardiab.biomedcentral.com/articles/10.1186/s12933-023-01854-z
Using AI to develop new brain-computer technology, stroke patients who have been aphasic for 18 years can "speak" again
A high-performance neuroprosthesis for speech decoding and avatar control
*Source : Nature
*Author : University of California team
*Interpretation : 18 years after she suffered a stroke and became speechless, AI + brain-computer interface helped her "voice her thoughts" (click to read the original text)
*paper :
A high-performance neuroprosthesis for speech decoding and avatar control | Nature
Commercial AI Lunit reads mammograms as accurately as doctors
Performance of a Breast Cancer Detection AI Algorithm Using the Personal Performance in Mammographic Screening Scheme
*Source : Radiology
*Author : Research team from the University of Nottingham, UK
*Interpretation : "Pink Killer" wanted notice, AI's ability to read mammograms is as good as that of doctors (click to read the original text)
*paper :
https://pubs.rsna.org/doi/10.1148/radiol.223299
Institute of Genomics, Chinese Academy of Sciences establishes open biomedical imaging archive
Self-supervised learning of hologram reconstruction using physics consistency
*Source : bioRxiv
*Author : Institute of Genomics, Chinese Academy of Sciences
*paper :
Retina image basic model RETFound predicts various systemic diseases
A foundation model for generalizable disease detection from retinal images
*Source : Nature
*Author : Zhou Yukun, PhD candidate at University College London and Moorfields Eye Hospital, and others
*paper :
A foundation model for generalizable disease detection from retinal images | Nature
Artificial intelligence detection of pancreatic cancer based on deep learning
Large-scale pancreatic cancer detection via non-contrast CT and deep learning
*Source : Nature Medicine
*Author : Alibaba Damo Hospital cooperates with many domestic and foreign medical institutions
*paper :
Large-scale pancreatic cancer detection via non-contrast CT and deep learning | Nature Medicine
Optimizing the design of triboelectric nanogenerator tactile sensors for text recognition and braille recognition
Machine Learning-Enabled Tactile Sensor Design for Dynamic Touch Decoding
*Source : Advanced Science
*Author : Research group of Yang Geng and Xu Kaichen from Zhejiang University
*Interpretation : Zhejiang University uses SVM to optimize tactile sensors, and the Braille recognition rate reaches 96.12% (click to read the original text)
*paper :
https://onlinelibrary.wiley.com/doi/10.1002/advs.202303949
AI+ Materials Chemistry
Combining multiple deep learning architectures to determine the internal structure of materials through surface observation
Fill in the Blank: Transferrable Deep Learning Approaches to Recover Missing Physical Field Information
*Source : Advanced Materials
*Author : Researchers at MIT
*Interpretation : "Fill in the blanks" in the material space: MIT uses deep learning to solve non-destructive testing problems (click to read the original text)
*paper :
https://onlinelibrary.wiley.com/doi/full/10.1002/adma.202301449
Combining deep neural networks and natural language processing to develop corrosion-resistant alloys
Enhancing corrosion-resistant alloy design through natural language processing and deep learning
*Source : Science Advances
*Author : Researchers from the Max Planck Institute for Iron Research in Germany
*Interpretation : AI "anti-corruption", Germany's Max Planck Institute combines NLP and DNN to develop anti-corrosion alloys (click to read the original text)
*paper :
https://www.science.org/doi/10.1126/sciadv.adg7992
Based on the machine learning model, AI is trained to extract structural parameters of porous materials to predict water adsorption isotherms.
Machine learning-assisted prediction of water adsorption isotherms and cooling performance
*Source : Journal of Materials Chemistry A
*Author : Li Song’s research group at Huazhong University of Science and Technology
*paper :
Field-induced recursive embedding of atomic neural network FIREANN accurately describes changes in external field intensity and direction
Universal machine learning for the response of atomistic systems to external fields
*Source : Nature Communication
*Author : Jiang Bin’s research group at University of Science and Technology of China
*paper :
DeepMind releases deep learning tool GNoME and discovers 2.2 million new crystals
Scaling deep learning for materials discovery
*Source : DeepMind
*Author : Nature
*Interpretation : 800 years ahead of mankind? DeepMind releases GNoME, using deep learning to predict 2.2 million new crystals (click to read the original text)
*paper :
Scaling deep learning for materials discovery | Nature
SEN machine learning model to achieve high-precision material performance prediction
Material symmetry recognition and property prediction accomplished by crystal capsule representation
*Source : Nature Communications
*Author : Research group of Li Huashan and Wang Biao of Sun Yat-sen University
*paper :
RetroExplainer algorithm performs retrosynthetic prediction based on deep learning
Retrosynthesis prediction with an interpretable deep-learning framework based on molecular assembly tasks
*Source : Nature Communications
*Author : Shandong University, University of Electronic Science and Technology of China research group
*paper:
Using machine learning to optimize cocatalysts for BiVO(4) photoanodes
A comprehensive machine learning strategy for designing high-performance photoanode catalysts
*Source : Journal of Materials Chemistry A
*Author : Zhu Hongwei’s research group at Tsinghua University
*paper :
AI+ animal and plant sciences
Population genetic method based on machine learning reveals the formation mechanism of grape flavor
Adaptive and maladaptive introgression in grapevine domestication
*Directory :Proceedings of the National Academy of Sciences
*Author : Researchers at Shenzhen Agrigenomics, Chinese Academy of Agricultural Sciences
*paper :
https://www.pnas.org/doi/abs/10.1073/pnas.2222041120
Use Python API and computer vision API to monitor the blooming of cherry blossoms in Japan
The spatiotemporal signature of cherry blossom flowering across Japan revealed via analysis of social network site images
*Source : Flora
*Author : Research team from Monash University, Australia
*paper :
https://www.sciencedirect.com/science/article/abs/pii/S0367253023001019
Review: Use AI to start bioinformatics research more efficiently
In addition to well-known bioinformatics developments like AlphaFold, AI has rich application cases in biological fields such as homology search, multiple alignment and phylogeny construction, genome sequence analysis, and gene discovery. As a biological researcher, being able to skillfully integrate machine learning tools into data analysis will definitely accelerate scientific discovery and improve scientific research efficiency.
*Recommended reading : Bioinformatics | Use AI to start research more efficiently (click to read the original text)
Deep learning method based on twin network to automatically capture the embryonic development process
Uncovering developmental time and tempo using deep learning
*Source : Nature Methods
*Author : Systems biologist Patrick Müller and researchers at the University of Konstanz
*Interpretation : AI combined with embryos? Systems biologist Patrick Müller uses twin networks to study zebrafish embryos (click to read the original text)
*paper :
Uncovering developmental time and tempo using deep learning | Nature Methods
Using more than 50,000 photos, train a multi-species image recognition model based on face recognition ArcFace Classification Head
A deep learning approach to photo–identification demonstrates high performance on two dozen cetacean species
*来源:Methods in Ecology and Evolution
*Author : Researchers at the University of Hawaii
*paper :
https://besjournals.onlinelibrary.wiley.com/doi/full/10.1111/2041-210X.14167
Using data from 628 Labrador retrievers and comparing 3 models, we found the behavioral characteristics that affect the performance of olfactory detection dogs.
Machine learning prediction and classification of behavioral selection in a canine olfactory detection program
*Source : Scientific Reports
*Author : Researchers at the Abigail Wexner Research Institute at Nationwide Children's Hospital and Rocky Vista University
*paper :
AI camera alert system accurately distinguishes tigers from other species
Accurate proteome-wide missense variant effect prediction with AlphaMissense
*Source : BioScience
*Author : Clemson University researchers
*Interpretation : To deal with the problem of coexistence of humans and tigers, the first AI camera to identify and transmit tiger photos is here (click to read the original text)
*paper :
https://www.science.org/doi/10.1126/science.adg7492
The BirdFlow model uses computer modeling and the eBird data set to accurately predict the flight paths of migratory birds
BirdFlow: Learning seasonal bird movements from eBird data
*来源:Methods in Ecology and Evolution
*Author : Researchers from Massachusetts State University and Cornell University
*Interpretation : With the help of computer modeling and eBird data set, the University of Massachusetts successfully predicted bird migration (click to read the original text)
*paper :
https://besjournals.onlinelibrary.wiley.com/doi/full/10.1111/2041-210X.14052
AI+ Agriculture, Forestry and Animal Husbandry
Using computer vision + deep learning to develop a cow lameness detection system with an accuracy of 94%-100%
Deep learning pose estimation for multi-cattle lameness detection
*Source : Nature
*Author : Researchers from Newcastle University and Ferra Sciences Ltd
*Interpretation : Leveraging computer vision and deep learning, Newcastle University develops a real-time, automated cow lameness detection system (click to read the original text)
*paper :
Deep learning pose estimation for multi-cattle lameness detection | Scientific Reports
Drone + AI image analysis to detect forestry pests
Testing early detection of pine processionary moth Thaumetopoea pityocampa nests using UAV-based methods
*Source : NeoBiota
*Author : University of Lisbon research team
*Interpretation : Drone + AI image analysis: University of Lisbon efficiently detects forestry pests (click to read the original text)
*paper :
Combining laboratory observations and machine learning, it is proven that ultrasonic waves emitted by tomato and tobacco plants under stressful conditions can propagate through the air.
Sounds emitted by plants under stress are airborne and informative
*Source : Cell
*Author : Researchers at Tel Aviv University, Israel
*Interpretation : Tomatoes can also "scream" under pressure. Tel Aviv University discovered that the plant kingdom is not silent (click to read the original text)
*paper :
Using the YOLOv5 algorithm, a model is designed to monitor the posture of sows and the birth of piglets.
Sow Farrowing Early Warning and Supervision for Embedded Board Implementations
*Source : Sensors
*Author : Research team of Nanjing Agricultural University
*Interpretation : We knew sows would give birth early, and this time Nannong used NVIDIA edge AI Jetson (click to read the original text)
*paper :
Use convolutional neural networks to quickly and accurately count rice yields
Deep Learning Enables Instant and Versatile Estimation of Rice Yield Using Ground-Based RGB Images
*Source : Plant Phenomics
*Author : Researchers at Kyoto University
*Interpretation : Kyoto University uses CNN to predict grain output. Don’t ask God for a good harvest, just ask AI (click to read the original text)
*paper:
https://spj.science.org/doi/10.34133/plantphenomics.0073
A systematic process for collecting plant phenotypic data using drones to predict the best harvest date
Drone-Based Harvest Data Prediction Can Reduce On-Farm Food Loss and Improve Farmer Income
*Source : Plant Phenomics
*Author : Researchers from the University of Tokyo and Chiba University
*Interpretation : Up to 20% of losses can be recovered! The University of Tokyo uses AI and drones to predict the best harvest dates for crops (click to read the original article)
*paper :
https://spj.science.org/doi/10.34133/plantphenomics.0086#body-ref-B4
AI+ meteorological research
The machine learning model CSU-MLP based on random forest accurately forecasts severe weather in the medium term (4-8 days)
A new paradigm for medium-range severe weather forecasts: probabilistic random forest-based predictions
*Source : Weather and Forecasting
*Author : Researchers from Colorado State University and National Oceanic and Atmospheric Administration
*paper :
https://arxiv.org/abs/2208.02383
Leveraging global storm analysis simulations and machine learning to create new algorithms to accurately predict extreme precipitation
Implicit learning of convective organization explains precipitation stochasticity
*Source : PNAS
*Author : Columbia University LEAP Laboratory
*Interpretation : Accurately predict extreme precipitation, Columbia University launches an upgraded version of the neural network Org-NN (click to read the original text)
*paper :
https://www.pnas.org/doi/10.1073/pnas.2216158120
Review: Collecting data from hailstorm centers and using large models to predict extreme weather
As early as 2021, Alibaba Cloud revealed that DAMO Academy and the National Meteorological Center jointly developed an AI algorithm for weather prediction and successfully predicted multiple severe convective weather. In September of the same year, Deepmind published an article in "Nature" using a deep generation model to conduct real-time rainfall forecasts.
At the beginning of 2023, Deepmind officially launched GraphCast, which can predict the global weather for the next 10 days with a resolution of 0.25° in one minute. In April, Nanjing University of Information Science and Technology and Shanghai Artificial Intelligence Laboratory jointly developed the "Fengwu" large-scale weather forecast model, which has further reduced errors compared to GraphCast.
Subsequently, Huawei launched the "Pangu" large-scale weather model. Due to the introduction of a three-dimensional neural network into the model, the prediction accuracy of "Pangu" exceeded the current most accurate NWP prediction system for the first time. Recently, Tsinghua University and Fudan University have successively released the “NowCastNet” and “Fuxi” models.
*Recommended reading : The Hailstorm Center collects data, large models support extreme weather forecasts, and "storm chasers" are taking place (click to read the original text)
Review: Data-driven machine learning weather forecasting models
Numerical weather prediction is the mainstream method of weather forecasting. It solves the state of the earth system grid by grid through numerical integration, which is a deductive reasoning process. Since 2022, machine learning models in the field of weather forecasting have achieved a series of breakthroughs, some of which can rival the high-precision forecasts of the European Center for Medium-Range Weather Forecasts.
*Recommended reading : Machine learning vs. numerical weather forecasting, how AI changes the existing weather forecasting model (click to read the original text)
AI+ Astronomy
Use simulated data to train computer vision algorithms to sharpen and "restore" astronomical images
Galaxy image deconvolution for weak gravitational lensing with unrolled plug-and-play ADMM
*Source : Monthly Notices of the Royal Astronomical Society
*Author : Research team from Tsinghua University and Northwestern University
*paper :
The PRIMO algorithm learns the light propagation rules around black holes and reconstructs clearer black hole images
The Image of the M87 Black Hole Reconstructed with PRIMO
*来源:The Astrophysical Journal Letters
*Author : Research team at the Institute for Advanced Study
*paper :
Using the unsupervised machine learning algorithm Astronomaly, we found anomalies that were previously overlooked.
Astronomaly at Scale: Searching for Anomalies Amongst 4 Million Galaxies
*Source : arXiv
*Author : Researcher at the University of the Western Cape
*Interpretation : Astronomaly: Using CNN and active learning to identify anomalies in 4 million galaxy images (click to read the original text)
*paper :
https://arxiv.org/abs/2309.08660
AI+ energy environment
Use machine learning to discover mineral combination patterns to predict mineral locations
Predicting new mineral occurrences and planetary analog environments via mineral association analysis
*Source : PNAS Nexus
*Author : Researchers at the Carnegie Institution for Science in Washington and the University of Arizona
*paper :
Using physical and machine learning models to predict pollution losses from the accumulation of dirt and other materials on the surface of photovoltaic panels in arid climates
Characterizing soiling losses for photovoltaic systems in dry climates: A case study in Cyprus
*Source : Solar Energy
*Author : Researchers at the University of Cyprus
*paper :
https://www.sciencedirect.com/science/article/pii/S0038092X23001883
Using machine learning methods to predict the emissions of harmful amine gases during the carbon capture process
Machine learning for industrial processes: Forecasting amine emissions from a carbon capture plant
*Source : ScienceAdvances
*Author : Research team composed of EPFL and Heriot-Watt University
*Interpretation : 8-year offensive campaign of post-90s academic masters: using machine learning to add BUFF to chemical research (click to read the original text)
*paper :
https://www.science.org/doi/10.1126/sciadv.adc9576
AI+ natural disasters
Stackable neural networks can analyze influencing factors in natural disasters
Landslide susceptibility modeling by interpretable neural network
*Source : Communications Earth & Environment
*Author : Researchers at UCLA
*Interpretation : The black box becomes transparent: UCLA develops an interpretable neural network SNN to predict landslides (click to read the original text)
*paper :
Using interpretable AI, we analyzed different geographical factors in Gippsland, Australia, and obtained the local wildfire probability distribution map.
Explainable artificial intelligence (XAI) for interpreting the contributing factors feed into the wildfire susceptibility prediction model
*Source : ScienceDirect
*Author : Researchers from Australian National University, University of Technology Sydney
*paper :
https://www.sciencedirect.com/science/article/pii/S0048969723016224
other
Direct and inverse problems of AI in meta-optics, data analysis based on metasurface systems
Artificial Intelligence in Meta-optics
*Source : ACS Publications
*Author : Researchers from City University of Hong Kong
*Interpretation : AI is increasing, and super optics has entered the era of rapid development (click to read the original text)
*paper :
https://pubs.acs.org/doi/10.1021/acs.chemrev.2c00012
Ithaca assists epigraphists in textual restoration, temporal and geographical attribution
Restoring and attributing ancient texts using deep neural networks
*Source : Nature
*Author : Researchers at DeepMind and University of Foscari in Venice
*Interpretation: New interpretation of the Millennium Code, DeepMind developed Ithaca to decipher Greek inscriptions (click to read the original text)
*paper :
Restoring and attributing ancient texts using deep neural networks | Nature
30 scholars jointly published a Nature review, a 10-year review deconstructing how AI reshapes the scientific research paradigm
Hanchen Wang, a postdoctoral fellow from the School of Computer Science and Gene Technology at Stanford University, Tianfan Fu from the Georgia Institute of Technology's Computational Science and Engineering Department, and Yuanqi Du from the Department of Computer Science at Cornell University, among other 30 people, reviewed the development of basic scientific research in the past ten years. The role of AI in , and raises the remaining challenges and shortcomings.
*Recommended reading : 30 scholars jointly published a Nature review, a 10-year review deconstructing how AI can reshape the scientific research paradigm (click to read the original text)
Policy: The Ministry of Science and Technology, together with the Natural Science Foundation of China, launched a special deployment of "Artificial Intelligence Driven Scientific Research" (AI for Science)
On March 27, Xinhua News Agency reported that in order to implement the national "New Generation Artificial Intelligence Development Plan", the Ministry of Science and Technology and the Natural Science Foundation of China have recently launched a special deployment of "Artificial Intelligence Driven Scientific Research" (AI for Science).
This time, my country has laid out an AI for Science cutting-edge technology research and development system that will closely integrate key issues in basic disciplines such as mathematics, physics, chemistry, astronomy, etc., and focus on scientific research needs in key areas such as drug research and development, genetic research, biological breeding, and new material research and development. In this regard, Xu Bo, director of the Institute of Automation of the Chinese Academy of Sciences, explained that fields such as new drug creation, genetic research, biological breeding, and new material research and development are urgently needed, outstanding, and representative directions for combining artificial intelligence with scientific research.
The above are the cutting-edge papers that "HyperAI Super Neural" will interpret for everyone in 2023. Follow the WeChat official account and reply to "2023 ScienceAI" in the background to package and download all papers.
In 2024, we will continue to pay attention to the latest scientific research results and related applications of AI for Science, so stay tuned~