Artificial Intelligence and Physics (Course No.580108)

Credits: 2

Class hours: 32

Language: Chinese-English bilingual

Assessment: Examination-free evaluation

Prerequisites: Advanced Mathematics, Linear Algebra and Probability Theory

Target students: Applied Physics, Optoelectronic Information Science and Engineering undergraduates.

Course description: With the rapid development of artificial intelligence, its methods are increasingly applied in natural-science research. This course introduces fundamental concepts of artificial intelligence, typical deep-learning frameworks and algorithms, and cutting-edge examples of AI for science. Students will practice applying these theories by completing group course projects.

Course objectives:

• Master basic methods for processing and analyzing practical datasets.
• Understand typical deep-learning frameworks and their correspondence to different data modalities.
• Gain insights into applying AI in physics and optoelectronic-engineering research.
• Become familiar with international frontiers of AI for science / engineering-intelligence research.
• Improve problem-solving capabilities and teamwork through group-based course projects.

Main content & class-hour allocation:

• Unit 1 Fundamentals of Artificial Intelligence (8 class hours)
  Machine-learning basics; Convolutional neural networks; Recurrent neural networks; Graph neural networks.
• Unit 2 Advanced Deep Learning (10 class hours)
  Optimization for deep learning; Reinforcement-learning basics; Transformer & large-language models; Generative-AI fundamentals (VAE, diffusion models); Programming practice with deep-learning frameworks.
• Unit3 AI applications in physics & project defense (14 class hours)
  Overview of AI-for-science; Case studies in theoretical physics, condensed-matter physics, acoustics, optoelectronics; Physical interpretability of deep learning; Group project presentation & defense.

Assessment: Course participation & written homework (40 %), final group project and defense (60 %).

Textbooks & references:
Main reference: Deep Learning by Ian Goodfellow, Yoshua Bengio, Aaron Courville.
Supplementary readings: Machine Learning (Zhou Zhihua); Statistical Learning Methods (Li Hang).