| Role | Name & Contact |
|---|---|
| Instructor | Renyu (Philip) Zhang Associate Professor, Department of Decisions, Operations and Technology, CUHK Business School π§ [email protected] π @911 Cheng Yu Tung Building |
| Teaching Assistant | Leo Cao Full-time TA, Department of Decisions, Operations and Technology, CUHK Business School π§ [email protected] |
| Tutorial Instructor | Xinyu Li PhD Candidate (Management Information Systems), CUHK Business School π§ [email protected] |
- π Website: https://github.com/rphilipzhang/AI-PhD-S25
- β° Time: Tuesday, 12:30pm-3:15pm (Jan 14 - Apr 15, 2025)
- Excluding: Jan 28 (Chinese New Year) and Mar 4 (Final Project Discussion)
- π Location: Wu Ho Man Yuen Building (WMY) 504
Welcome to the mono-repo of DOTE 6635: AI for Business Research at CUHK Business School!
- π§ Gain fundamental understanding of ML/AI concepts and methods relevant to business research.
- π‘ Explore applications of ML/AI in business research over the past decade.
- π Discover and nuture the taste of cutting-edge AI/ML technologies and their potential in your research domain.
Need to join remotely? Use our Zoom link (please seek approval from Philip):
- π₯ Join Meeting
- Meeting ID: 918 6344 5131
- Passcode: 459761
Most of the code in this course will be distributed through the Google CoLab cloud computing environment to avoid the incompatibility and version control issues on your local individual computer. On the other hand, you can always download the Jupyter Notebook from CoLab and run it your own computer.
- π The CoLab files of this course can be found at this folder.
- π The Google Sheet to sign up for groups and group tasks can be found here.
- π The overleaf template for scribing the lecture notes of this course can be found here.
- π¬ The replication projects can be found here.
- π₯οΈ The HPC Server compute resource of the CUHK DOT Department can be found here.
If you have any feedback on this course, please directly contact Philip at [email protected] and we will try our best to address it.
- ποΈ GitHub Repos: Spring 2024@CUHK, Summer 2024@SJTU Antai
- π₯ Video Recordings (You need to apply for access): Spring 2024@CUHK, Summer 2024@SJTU Antai
- π Scribed Notes: Spring 2024@CUHK
Subject to modifications. All classes start at 12:30pm and end at 3:15pm.
| Session | Date | Topic | Key Words |
|---|---|---|---|
| 1 | 1.14 | AI/ML in a Nutshell | Course Intro, ML Models, Model Evaluations |
| 2 | 1.21 | Intro to DL | DL Intro, Neural Nets, Prediction in Biz Research |
| 3 | 2.04 | LLM (I) | Transformer, Mamba |
| 4 | 2.11 | LLM (II) | Pre-training, BERT, GPT, Quantization |
| 5 | 2.18 | LLM (III) | Post-training, Fine-tuning, RLHF, Inference |
| 6 | 2.25 | LLM (IV) | Agentic AI, AI as Human Simulators |
| 7 | 3.04 | LLM (IV.V) | Final Project Discussions, Applications in Business Research |
| 8 | 3.11 | Causal (I) | Causal Inference Intro |
| 9 | 3.18 | Causal (II) | ML-Powered Causal Inference, Causal Trees and Forests |
| 10 | 3.25 | Causal (III) | Double Machine Learning, Neyman Orthogonality |
| 11 | 4.01 | Causal (IV) | RL, Off-Policy Evaluation |
| 12 | 4.08 | Economics and Ethics of AI | Economic Impact, Fairness, Bias of AI |
| 13 | 4.15 | Final Project Presentation | Discussions and Course Wrap-up |
All problem sets are due at 12:30pm right before class.
| Date | Time | Event | Note |
|---|---|---|---|
| 1.15 | 11:59pm | Group Sign-Ups | Each group has at most two students. |
| 1.17 | 7:00pm-9:00pm | Python Tutorial | Given by Xinyu Li, Python Tutorial CoLab |
| 1.24 | 7:00pm-9:00pm | PyTorch and DOT HPC Server Tutorial | Given by Xinyu Li, PyTorch Tutorial CoLab |
| 3.04 | 9:00am-6:00pm | Final Project Discussion | Please schedule a meeting with Philip. |
| 3.11 | 12:30pm | Final Project Proposal | 1-page maximum |
| 4.30 | 11:59pm | Scribed Lecture Notes | Overleaf link |
| 5.11 | 11:59pm | Project Paper, Slides, and Code | Paper page limit: 10 |
Find more on the Syllabus.
-
π Books:
-
π Courses:
- Foundations: ML Intro by Andrew Ng, DL Intro by Andrew Ng, Generative AI by Andrew Ng, Introduction to Causal Inference by Brady Neal
- Advanced Technologies: NLP (CS224N) by Chris Manning, CV (CS231N) by Fei-Fei Li, Deep Unsupervised Learning by Pieter Abbeel, DLR by Sergey Levine, DL Theory by Matus Telgarsky, LLM by Danqi Chen, Efficient Deep Learning Computing by Song Han, Deep Generative Models by Kaiming He, LLM Agents by Dawn Song, Advanced LLM Agents by Dawn Song, Data, Learning, and Algorithms by Tengyuan Liang, Hands-on DRL by Weinan Zhang, Jian Shen, and Yu Yong (in Chinese), Understanding LLM: Foundations and Safety by Dawn Song
- Biz/Econ Applications of AI: Machine Learning and Big Data by Melissa Dell and Matthew Harding, Digital Economics and the Economics of AI by Martin Beraja, Chiara Farronato, Avi Goldfarb, and Catherine Tucker, Generative AI and Causal Inference with Texts, NLP for Computational Social Science by Diyi Yang
-
π‘ Tutorials and Blogs:
- GitHub of Andrej Karpathy, Blog of Lilian Weng, Double Machine Learning Package Documentation, Causality and Deep Learning (ICML 2022 Tutorial), Causal Inference and Machine Learning (KDD 2021 Tutorial), Online Causal Inference Seminar, Training a Chinese LLM from Scratch (in Chinese), Physics of Language Models (ICML 2024 Tutorial), Counterfactual Learning and Evaluation for Recommender Systems: Foundations, Implementations, and Recent Advances (RecSys 2021 Tutorial), Language Agents: Foundations, Prospects, and Risks (EMNLP 2024 Tutorial), GitHub Repo: Upgrading Cursor to Devin
The following schedule is tentative and subject to changes.
- π Keywords: Course Introduction, Prediction in Biz Research, Basic ML Models
- π Slides: Course Intro, Prediction, ML Intro
- π» CoLab Notebook Demos: Bootstrap, k-Nearest Neighbors, Decision Tree, Random Forest, Gradient Boosting Tree
- βοΈ Homework: Problem Set 1 - Housing Price Prediction, due at 12:30pm, Feb/4/2025
- π Online Python Tutorial: Python Tutorial CoLab, 7:00pm-9:00pm, Jan/17/2025 (Friday), given by Xinyu Li, [email protected]. Zoom Link, Meeting ID: 939 4486 4920, Passcode: 456911
- π References:
- The Elements of Statistical Learning (2nd Edition), 2009, by Trevor Hastie, Robert Tibshirani, Jerome Friedman, link to ESL.
- Probabilistic Machine Learning: An Introduction, 2022, by Kevin Murphy, link to PML.
- Mullainathan, Sendhil, and Jann Spiess. 2017. Machine learning: an applied econometric approach. Journal of Economic Perspectives 31(2): 87-106.
- Athey, Susan, and Guido W. Imbens. 2019. Machine learning methods that economists should know about. Annual Review of Economics 11: 685-725.
- Kleinberg, Jon, Jens Ludwig, Sendhil Mullainathan, and Ziad Obermeyer. 2015. Prediction policy problems. American Economic Review 105(5): 491-495.
- Hofman, Jake M., et al. 2021. Integrating explanation and prediction in computational social science. Nature 595.7866: 181-188.
- Bastani, Hamsa, Dennis Zhang, and Heng Zhang. 2022. Applied machine learning in operations management. Innovative Technology at the Interface of Finance and Operations. Springer: 189-222.
- Kelly, Brian, and Dacheng Xiu. 2023. Financial machine learning, SSRN, link to the paper.
- The Bitter Lesson, by Rich Sutton, which develops so far the most critical insight of AI: "The biggest lesson that can be read from 70 years of AI research is that general methods that leverage computation are ultimately the most effective, and by a large margin."
- Chatpers 1 & 3.2, Scribed Notes of Spring 2024 Course Offering.
- π Keywords: Bias-Variance Trade-off, Cross Validation, Bootstrap, Neural Nets, Computational Issues of Deep Learning
- π Slides: ML Intro, DL Intro
- π» CoLab Notebook Demos: Gradient Descent, Chain Rule, He Innitialization
- βοΈ Homework: Problem Set 2: Implementing Neural Nets, due at 12:30pm, Feb/11/2025
- π Online PyTorch and DOT HPC Server Tutorial: PyTorch Tutorial CoLab, 7:00pm-9:00pm, Jan/24/2025 (Friday), given by Xinyu Li, [email protected]. Zoom Link, Meeting ID: 939 4486 4920, Passcode: 456911
- π References:
- Deep Learning, 2016, by Ian Goodfellow, Yoshua Bengio and Aaron Courville, link to DL.
- Dive into Deep Learning (2nd Edition), 2023, by Aston Zhang, Zack Lipton, Mu Li, and Alex J. Smola, link to d2dl.
- Probabilistic Machine Learning: Advanced Topics, 2023, by Kevin Murphy, link to PML2.
- Deep Learning with PyTorch, 2020, by Eli Stevens, Luca Antiga, and Thomas Viehmann.
- Dell, Mellissa. 2024. Deep learning for economists. Journal of Economic Literature, forthcoming, link to the paper.
- Davies, A., VeliΔkoviΔ, P., Buesing, L., Blackwell, S., Zheng, D., TomaΕ‘ev, N., Tanburn, R., Battaglia, P., Blundell, C., JuhΓ‘sz, A. and Lackenby, M., 2021. Advancing mathematics by guiding human intuition with AI. Nature, 600(7887), pp.70-74.
- Ye, Z., Zhang, Z., Zhang, D., Zhang, H. and Zhang, R.P., 2023. Deep-Learning-Based Causal Inference for Large-Scale Combinatorial Experiments: Theory and Empirical Evidence. Available at SSRN 4375327, link to the paper.
- Luyang Chen, Markus Pelger, Jason Zhu (2023) Deep Learning in Asset Pricing. Management Science 70(2):714-750.
- Wang, Z., Gao, R. and Li, S. 2024. Neural-Network Mixed Logit Choice Model: Statistical and Optimality Guarantees. Working paper.
- Why Does Adam Work So Well? (in Chinese), Overview of gradient descent algorithms
- Chatpers 1 & 2, Scribed Notes of Spring 2024 Course Offering.
- π Keywords: Deep Learning Computations, Seq2Seq, Attention Mechanism, Transformer
- π Slides: What's New, DL Intro, Transformer
- π» CoLab Notebook Demos: Dropout, Micrograd, Attention Mechanism
- βοΈ Homework: Problem Set 2: Implementing Neural Nets, due at 12:30pm, Feb/11/2025
- π Presentation of Replication Project: By Jiaci Yi and Yachong Wang
- Gui, G. and Toubia, O., 2023. The challenge of using LLMs to simulate human behavior: A causal inference perspective. arXiv:2312.15524. Link to the paper.
- π References:
- Deep Learning, 2016, by Ian Goodfellow, Yoshua Bengio and Aaron Courville, link to DL.
- Dive into Deep Learning (2nd Edition), 2023, by Aston Zhang, Zack Lipton, Mu Li, and Alex J. Smola, link to d2dl.
- Dell, Mellissa. 2024. Deep learning for economists. Journal of Economic Literature, forthcoming, link to the paper.
- Sutskever, Ilya, Oriol Vinyals, and Quoc V. Le. 2014. Sequence to sequence learning with neural networks. Advances in neural information processing systems, 27.
- Lecture Notes and Slides for CS224n: Natural Language Processing with Deep Learning, by Christopher D. Manning, Diyi Yang, and Tatsunori Hashimoto. Link to CS224n.
- Parameter Initialization and Batch Normalization (in Chinese), GPU Comparisons, GitHub Repo for Micrograd by Andrej Karpathy.
- RNN and LSTM Visualizations, PyTorch's Tutorial of Seq2Seq for Machine Translation.
- Chatpers 2 & 6, Scribed Notes of Spring 2024 Course Offering.
- Handwritten Notes
- π Keywords: Transformer, Pretraining, BERT, GPT
- π Slides: What's New, Transformer, Pretraining
- π» CoLab Notebook Demos: Attention Mechanism, Transformer, BERT API @ Hugging Face
- βοΈ Homework: Problem Set 3: Sentiment Analysis with BERT, due at 12:30pm, Feb/25/2025
- π Presentation of Replication Project: By Xiqing Qin and Yuxin Chen
- Manning, B.S., Zhu, K. and Horton, J.J., 2024. Automated social science: Language models as scientists and subjects (No. w32381). National Bureau of Economic Research. Link to the paper, link to GitHub Repo.
- π References:
- Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., ... and Polosukhin, I. (2017). Attention is all you need. Advances in neural information processing systems, 30.
- Devlin, Jacob, Ming-Wei Chang, Kenton Lee, Kristina Toutanova. 2018. BERT: Pre-training of deep bidirectional transformers for language understanding. ArXiv preprint arXiv:1810.04805. GitHub Repo
- Radford, Alec, Karthik Narasimhan, Tim Salimans, and Ilya Sutskever. 2018. Improving language understanding by generative pre-training, (GPT-1) PDF link, GitHub Repo
- Radford, Alec, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, Ilya Sutskever. 2019. Language models are unsupervised multitask learners. OpenAI blog, 1(8), 9. (GPT-2) PDF Link, GitHub Repo
- Brown, Tom, et al. 2020. Language models are few-shot learners. Advances in neural information processing systems, 33, 1877-1901. (GPT-3) GitHub Repo
- Huang, Allen H., Hui Wang, and Yi Yang. 2023. FinBERT: A large language model for extracting information from financial text. Contemporary Accounting Research, 40(2): 806-841. GitHub Repo
- Chapter 11, Dive into Deep Learning (2nd Edition), 2023, by Aston Zhang, Zack Lipton, Mu Li, and Alex J. Smola, link to d2dl.
- Lecture Notes and Slides for CS224n: Natural Language Processing with Deep Learning, by Christopher D. Manning, Diyi Yang, and Tatsunori Hashimoto. Link to CS224n.
- Part 2 & 4, Slides for COS 597G: Understanding Large Language Models, by Danqi Chen. Link to COS 597G
- Illustrated Transformer, Transformer from Scratch with the Code on GitHub.
- A Visual Guide to BERT, How GPT-3 Works
- Andrej Karpathy's Lectures: Build GPT-2 (124M) from Scratch, Deep Dive into LLM
- Chatpers 7 & 8 Scribed Notes of Spring 2024 Course Offering.