Fall Semester, 2026 (101 Sherred).  Attendance Check-in
TT 10:40am - 12:00pm 

Reference Books

Reference Books	Title and Author
	Chen, Y., Chen, Y., Fan, J., and Ma, C. (2021). Spectral Methods for Data Science: A Statistical Perspective. Foundations and Trends in Machine Learning.
	Fan, J., Li, R., Zhang, C.-H., and Zou (2020).  Statistical Foundations of Data Science. CRC Press.

General Information

Instructor: Jianqing Fan, Frederick L. Moore'18 Professor of Finance.
Office: 205 Sherred Hall
Phone: 258-7924
E-mail: [email protected]
Office Hours: Tuesday 1:45 pm –- 2:45 pm, Thursday 9:30 am--10:30 am (205 Sherred), or by appointment.

Reference Books

• Fan, J., Li, R., Zhang, C.-H., and Zou, H. (2020). Statistical Foundations of Data Science. CRC Press. (Chapters 9--11)
• Chen, Y., Chen, Y., Fan, J., and Ma, C. (2021).Spectral Methods for Data Science: A Statistical Perspective. Foundations and Trends in Machine Learning.

Syllabus

This course covers key topics in statistical machine learning and generative AI, with neural networks and transformers as core modeling backbones. Topics include: (1) Factor-adjusted nonparametric Lasso; (2) causality pursuit and causal inference; (3) generative models; (4) fine tuning and preference modeling; (5) large language models and societies; (6) transformers and in context learning; and (7) additional topics. Students are expected to participate actively in paper surveys and presentations.

The course will cover the following topics.

1. Introduction to Deep Learning and Generative AI (3 Lectures)
   • FNN, ResNet, CNN, RNN
   • Transformers & LLMs
   • Watermarking
   • Generative Adversarial Nets
   • Diffusion Models
   • Flow Matching
   • Training algorithms
2. Fine-tuning, Low-rank Matrices, and Their Applications (2 Lectures)
   • Fine-tune LLMs
   • Preference Data and Item Ranking
   • Network Data and Community Detection
   • Topic Modeling
   • Matrix Completion
   • Factor Models and Regularization
3. Neural Networks in Statistical Learning and Inference (4 Lectures)
   • Neural Approximation Theory
   • Attribution Learning:  Factor-adjusted Nonparametric Variable Selection
   • Causal Inference
   • Causality Learning
   • Transfer Learning with Attributions
   • Semiparametric Learning
4. AI for Society (2 Lectures)
   • Financial Statement Fraud Detections
   • Measuring Misinformation
   • Auditing Corporate Narratives
5.  Generative AI (2 Lectures)
   • Score Matching and Flow Matching
   • In-context conditional data generation
   • Statistical inference via conditional generation
6. Other Topics: Papers of Students' Choice

Attendance

Attendance of the class is required and essential.  The course materials are mainly from the notes.  

Schedules and Tentative Grading Policy