Artificial Intelligence for Science

The following sections are included:

• About the Editors
• Contents

The following sections are included:

• A Brief History of Scientific Discovery
• Artificial Intelligence, Machine Learning, and Deep Neural Networks
• AI and Modern Scientific Discovery
• AI for Science Book
• Summary
• References

The following sections are included:

• Overview
• Setting the Scene: Chapters 3–10
• Exploring Application Domains: Chapters 11–26
• The Ecosystem of AI for Science: Chapters 27–33
• Perspectives on AI for Science: Chapters 34–38
• Endpiece: AI Methods and the Concepts Used in the Book — Chapters 39 and 40

The following sections are included:

• Introduction
• Discovering Convergent Patterns
• Turning Patterns into Formal Processes
• Design Principles for Scalability and Computability of Large Datasets
• Economies of Scale — From SkyServer to SciServer
• Open Science and Large Open Data Collections
• Long-Term High-Value Datasets
• The Path Ahead
• Acknowledgments
• References

The following sections are included:

• Introduction
• A Brief History of Data Science — A US Perspective
• The 4 + 1 Model of Data Science
• AI from the Perspective of the 4 + 1 Model
• Understandable AI
• Conclusion
• References

The following sections are included:

• Introduction
• Background
• Results
• How Did DeepMind Win CASP?
• The AlphaFold Protein Structure Database²
• Did AlphaFold Actually Solve the Protein Folding Problem?
• Implications for Experimental Structural Biologists
• References

The following sections are included:

• Introduction and Background
• Early Applications: Digital Sky Surveys
• Increasing Challenges: Time-Domain Astronomy
• A Growing Applications Landscape
• Concluding Comments and Future Prospects
• References

The following sections are included:

• Introduction
• Machine Learning Challenges and Requirements
• Optimizing Instrument Operational Performance
• Optimizing Instrument Design
• Conclusion
• References

The following sections are included:

• The Rapid Evolution of Experimental Facilities
• What is “Artificial Intelligence”?
• Challenges in Applying AI/ML to Experimental Science
• Case Studies
• Future Opportunities for AI/ML in Experiment and Computation
• AI/ML for Discovering and Understanding the Scientific Landscape
• References

The following sections are included:

• Fugaku: The State-of-the-art CPU-based Supercomputer for Driving AI-for-Science
• AI-for-Science Applications — Case Studies
• Architectures for AI-for-Science — Past, Present, and Future
• References

The following sections are included:

• Introduction
• Role of AI in Science
• Approaches and Benefits
• Future Directions
• Acknowledgments
• References

The following sections are included:

• Modern Radio Astronomy and the Square Kilometre Array
• The Big Data Challenge of the SKA: Why AI is Needed
• AI for Image Analysis in Radio Astronomy
• AI for Time-domain Analysis in Radio Astronomy
• AI Challenges for the SKA
• References

The following sections are included:

• Introduction
• The Emergence of Machine Learning in Astronomy
• Open Data and Open Science
• The Democratization of Machine Learning with the Emergence of Open Frameworks
• The Challenge of Machine Learning for Streaming Data
• Physics-Driven Approaches to Machine Learning
• Scalability of Machine Learning
• Training and Software Engineering in the Sciences
• The Challenges and Opportunities Ahead
• References

The following sections are included:

• Introduction
• AI for Net-Zero
• Net-Zero for AI
• Conclusion and Vision for Future
• References

The following sections are included:

• Introduction
• Climate Modeling
• Analysis of Climate Model Data
• AI for Earth Observations in Climate Science
• How AI/ML Will Transform Climate Science
• References

The following sections are included:

• Introduction
• Challenges in Fusion Reactor Operations and Control
• Acceleration and Validation of Simulation
• Design Optimization
• AI and the Future of Fusion
• Acknowledgments
• References

The following sections are included:

• Introduction
• Virtualization of Power Grid Operations
• Human–Machine Collaboration for Efficient Grid Operations
• Distributed Energy Resources for Efficient Grid Services
• Summary
• References

The following sections are included:

• Introduction
• AI and Machine Learning in Remote Sensing
• Objectives for Observations
• Radiative Transfer, AI, and Machine Learning
• Examples of Machine Learning in Interpreting Remotely Sensed Imagery
• Concluding Remarks
• References

The following sections are included:

• Introduction
• Prospects and Promises
• Biological Challenges
• Governance Challenges
• Data Challenges
• Conclusion: AI as Engagement Platform for Agricultural Development
• Competing Interests
• Grant Information
• References

The following sections are included:

• Introduction
• Digital Histopathology: A Brief Background
• Applications of AI in Digital Histopathology
• Conclusions
• References

The following sections are included:

• Introduction
• Modeling Approaches
• Applications
• Decision Support Tools
• Discussion and Concluding Remarks
• Acknowledgments
• References

The following sections are included:

• Introduction
• Traditional, “Narrow”, AI
• How Far Can We Go with AI?
• The Problem with Traditional AI
• The Rise of Big AI
• Coping with Chaos
• Big AI in Drug Discovery
• Big AI in the Virtual Human
• The Future of Big AI
• Acknowledgment
• References

The following sections are included:

• Introduction
• Challenges
• Emerging Solutions
• Opportunities
• Conclusions
• References

The following sections are included:

• Introduction
• AI Allows You to Make a Material Model
• Application Modalities
• Autonomous Science
• Challenges
• Conclusion and Outlook
• References

The following sections are included:

• Introduction
• How Can AI Advance Materials Science and Engineering?
• AI for Materials: Challenges and Opportunities
• AI for Materials: Illustrative Applications
• AI for Materials: Outlook and Broader Impact
• Summary
• Acknowledgments
• References

The following sections are included:

• Introduction
• Particle-level Inference
• Event-level Inference
• Experiment-level Inference
• Simulation
• Conclusion
• References

The following sections are included:

• Introduction
• Surrogate Models: Looking for a Needle in a Haystack
• ML Models as Fitting Functions
• Multidimensional Integration Using Machine Learning
• Lattice QCD
• ML Methods for Efficient Generation of Gauge Configurations: Application to Lattice QCD
• Contrasting Two Black Boxes: Lattice QCD and ML
• A General Method for Bias Correction in ML Predictions of Averages
• Examples of Physics-aware Machine Learning Models
• Unsupervised ML Approaches
• Exploring the Space of Possible Theories
• Interpreting Extra Dimensions from Deep Neural Nets
• Conclusions
• Acknowledgments
• References

The following sections are included:

• Introduction
• Enhanced Web Search with Schema.org
• Dataset Search: A Dedicated Search Portal
• Bioschemas: Schema.org for the Life Sciences
• Applications Consuming Bioschemas
• Conclusions and Future Research Directions
• Acknowledgments
• References

The following sections are included:

• Introduction
• Learning Everywhere Paradigm
• Learning Everywhere Examples
• Machine Learning and Scientific Workflow Applications
• Discussion
• Acknowledgments
• References

The following sections are included:

• Introduction
• Specific Challenges
• Scientific Visualization and AI
• DNN Use Case: Visualization Surrogates
• References

The following sections are included:

• Introduction
• Review of UQ in DL
• Separating the Wheat from the Chaff
• Closing Remarks
• References

The following sections are included:

• Introduction
• Innovations Brought on by AI
• Open Research Challenges
• Statistics from Real Networks: Case Study from US Network (ESnet)
• Background: Current and Emerging Networking Technologies
• A WAN Traffic Prediction Case Study
• Towards Building Self-driving Networks
• Outlook and Next Steps
• Conclusion
• References

The following sections are included:

• Introduction
• Bayesian Approach
• Prior Construction
• Decision-Making under Uncertainty
• Optimal Experimental Design
• Conclusion: From Refining Probabilities to Expanding Possibilities
• Acknowledgments
• References

The following sections are included:

• The FAIR Principles: A Prerequisite for Data Science
• Machine-Readable Metadata Matters
• FAIR Evaluation: Where Are We?
• FAIR Data Strategy in Industry
• Moving Forward
• References

The following sections are included:

• Introduction
• From Natural to Scientific Language
• Opportunities and Open Problems
• Conclusion
• References

The following sections are included:

• Introduction
• Sensors
• AV System Architecture
• Perception
• Conclusion
• References

The following sections are included:

• Introduction
• AI-led Automation in Science
• Key Milestones
• The Future of AI-led Discovery in Science
• The Evaluation of AI in Science
• Conclusion
• References

The following sections are included:

• Introduction: Eighty Years of AI for Science
• The Imperative of AI for Science
• General Intelligent Capabilities of AI Systems for Science: Thoughtful AI
• Specific Capabilities of AI Systems for Science: Core Competencies
• Reflection Competency in Intelligent Systems for Science
• Conclusion
• Acknowledgments
• References

The following sections are included:

• Introduction
• Explaining the Predictions of Deep Neural Networks
• Inferring a Nonlinear Causal Relationship with a Continuous Treatment
• Conclusion
• References

The following sections are included:

• Introduction
• Classical AI/Machine Learning
• Deep Learning Components
• Deep Learning Networks
• References

The following sections are included:

• Overview
• Chapter 3. Data-Driven Science in the Era of AI: From Patterns to Practice
• Chapter 4. AI in the Broader Context of Data Science
• Chapter 5. AlphaFold — The End of the Protein Folding Problem or the Start of Something Bigger?
• Chapter 6. Applications of AI in Astronomy
• Chapter 7. Machine Learning for Complex Instrument Design and Optimization
• Chapter 8. Artificial Intelligence (AI) and Machine Learning (ML) at Experimental Facilities
• Chapter 9. The First Exascale Supercomputer: Accelerating AI-for-Science and Beyond
• Chapter 10. Benchmarking for AI for Science
• Chapter 11. Radio Astronomy and the Square Kilometer Array
• Chapter 12. The Rise of the Machines
• Chapter 13. AI for Net-Zero
• Chapter 14. AI for Climate Science
• Chapter 15. Accelerating Fusion Energy with AI
• Chapter 16. Artificial Intelligence for a Resilient and Flexible Power Grid
• Chapter 17. AI and Machine Learning in Observing Earth from Space
• Chapter 18. Artificial Intelligence in Plant and Agricultural Research
• Chapter 19. AI and Pathology: Steering Treatment and Predicting Outcomes
• Chapter 20. The Role of Artificial Intelligence in Epidemiological Modeling
• Chapter 21. Big AI: Blending Big Data with Big Theory to Build Virtual Humans
• Chapter 22. A Roadmap for Defining Machine Learning Standards in Life Sciences
• Chapter 23. Artificial Intelligence for Materials
• Chapter 24. Artificial Intelligence for Accelerating Materials Discovery
• Chapter 25. Experimental Particle Physics and Artificial Intelligence
• Chapter 26. AI and Theoretical Particle Physics
• Chapter 27. Schema.org for Scientific Data
• Chapter 28. AI-Enabled HPC Workflows
• Chapter 29. AI for Scientific Visualization
• Chapter 30. Uncertainty Quantification in AI for Science
• Chapter 31. AI for Next-Generation Global Network-Integrated Systems and Testbeds
• Chapter 32. AI for Optimal Experimental Design and Decision-Making
• Chapter 33. FAIR: Making Data AI-Ready
• Chapter 34. Large Language Models for Science
• Chapter 35. Autonomous Vehicles
• Chapter 36. The Automated AI-Driven Future of Scientific Discovery
• Chapter 37. Towards Reflection Competencies in Intelligent Systems for Science
• Chapter 38. The Interface of Machine Learning and Causal Inference

The following section is included:

• Index