This interdisciplinary course, AI in Science, introduces the foundational concepts and practical tools of Artificial Intelligence (AI) tailored for students in the sciences. Co-taught by faculty from various departments within the College of Science, the course explores how AI is transforming scientific research and discovery across disciplines such as physics, mathematics, psychology, and earth sciences. Students will gain an understanding of key AI techniques—including machine learning, data analysis, and modeling—and learn how these tools can be applied to real-world scientific problems. The course combines conceptual lectures with hands-on sessions, equipping students with both theoretical insights and practical skills to begin incorporating AI into their academic and research pursuits. No prior experience in AI or computer science is required. This course is designed as a starting point for science students to engage with AI in meaningful and discipline-relevant ways. 欲索取加選授權碼者，請填以下表單：https://forms.gle/hfiaP1MbtQXmryir6

By the end of this course, students will be able to: 1. Understand fundamental concepts of Artificial Intelligence and their relevance to various scientific disciplines. 2. Recognize key AI techniques such as machine learning, data-driven modeling, and statistical inference, and how they are applied in scientific research. 3. Explore real-world case studies demonstrating the use of AI in fields like physics, mathematics, psychology, and earth sciences. 4. Gain familiarity with commonly used AI tools and programming environments (e.g., Python, Jupyter Notebooks, scikit-learn, etc.). 5. Develop basic skills to analyze scientific data using AI-driven approaches. 6. Formulate potential applications of AI in their own areas of study or research. 7. Collaborate across disciplines to discuss and solve scientific problems using AI techniques.

No prior experience in AI or computer science is required. This course is designed as a starting point for science students to engage with AI in meaningful and discipline-relevant ways.

2

These readings are selected to provide all students—regardless of department—with a common foundation in AI concepts and practical tools: 1. Artificial Intelligence: A Guide for Thinking Humans – Melanie Mitchell o An accessible and thoughtful overview of AI fundamentals, suitable for students from all scientific backgrounds. o Helps establish conceptual understanding before diving into domain-specific applications. 2. Python Data Science Handbook – Jake VanderPlas (Selected chapters) o Practical guide to tools commonly used in scientific AI: NumPy, pandas, matplotlib, and scikit-learn. o Provides a technical base for students to complete hands-on exercises and apply AI in their fields. 3. Lecture Slides and Faculty Notes (Course Pack) o Custom-written notes and slide decks prepared by co-lecturing professors in physics, mathematics, psychology, and earth sciences. o Focuses on domain-relevant applications of AI and problem-solving strategies.

These optional readings deepen students' understanding of AI applications in specific disciplines: Physics & Mathematics • Data-Driven Science and Engineering: Machine Learning, Dynamical Systems, and Control – Steven L. Brunton & J. Nathan Kutz A math-forward book ideal for students interested in modeling physical systems using AI and control theory. • Relevant Research Articles o Examples: AI in high-energy physics, pattern discovery in large datasets, symbolic regression for physical law discovery. Psychology • The Book of Why: The New Science of Cause and Effect – Judea Pearl Offers insight into causal inference and its importance in psychological and behavioral sciences. • Selected Readings from Cognitive Science Journals o Topics: AI models of human cognition, neural networks vs. brain function, predictive modeling in psychology. Earth Sciences • Review Articles from Journals like: o Nature Geoscience, Earth and Space Science, Computers & Geosciences o Topics include remote sensing, environmental modeling, and climate prediction using machine learning. • Case Studies o Use of deep learning for earthquake prediction, land-use classification, or atmospheric modeling.