Lecture Slides

© Zhigang Feng — shared for personal study. Every deck is free to download; no password needed. Each page carries a copyright watermark. Please do not redistribute, repost, or mirror them without permission, and cite the source if you build on them.

Ten lectures, split into short topic decks so each one can be read in a sitting. The syllabus says which decks belong to which module, and the labs page has the matching notebooks.

Lecture 1 — Introduction: AI and Economic Research Available

Three topic decks.

T1 — AI & Economics Download PDF
T2 — AI as a Research Collaborator Download PDF
T3 — Course, Tools & Projects Download PDF

Lecture 2 — What is AI? Available

Two topic decks.

T1 — ML & the Predictive Stack Download PDF
T2 — Generative & Agentic AI Download PDF

Lecture 3 — Machine Learning for Quantitative Macroeconomists Available

Four topic decks.

T1 — Why ML & Pattern Recognition Download PDF
T2 — Neural-Network Concepts Download PDF
T3 — Training & Backpropagation Download PDF
T4 — Specialized Networks & Lab (CNN, ICNN, monotone nets) Download PDF

Lecture 4 — Solving Macro Models via Deep Learning & RL Available

Four topic decks.

T1 — Motivation & the Optimal-Growth Model Download PDF
T2 — Deep VFI & Actor-Critic Download PDF
T3 — Euler-Equation Method & PyTorch Download PDF
T4 — Structured NNs for Equilibrium & Validation Download PDF

Lecture 5 — Reinforcement Learning in a Nutshell Available

Three topic decks — one map, three acts: Model, Sample, Scale.

T1 — RL Framework & Dynamic Programming Download PDF
T2 — Monte Carlo, Exploration & Temporal-Difference Learning Download PDF
T3 — Actor-Critic, Deep RL & Applications Download PDF

Lecture 7 — Large Language Models & Text as Economic Data Available

Five topic decks — measure, represent, attend.

T1 — Text as Economic Measurement Download PDF
T2a — Preprocessing & Bag-of-Words Download PDF
T2b — Word Embeddings Download PDF
T3a — The Attention Mechanism Download PDF
T3b — The Transformer Download PDF

Lecture 8 — Retrieval-Augmented Generation (RAG) Available

Five topic decks — one map, five acts: the wall, index, retrieve, generate, apply. T1–T3 build the pipeline; T4 turns retrieved evidence into a cited answer; T5 puts it to work on real research corpora.

T1 — The Wall: Why Naive Prompting Fails Download PDF
T2 — Index: Chunking & Embeddings Download PDF
T3 — Retrieve: Vector Search & GraphRAG Download PDF
T4 — Generate: Grounded Answers, and RAG vs. Fine-Tuning Download PDF
T5 — Apply: Research Corpora, Demo & Evaluation Download PDF

Lecture 9 — Agentic AI for Research Workflows Available

Five topic decks — one map, five acts: Framing, Machine, Method, Management, Mindset.

T1 — Framing: What Agentic AI Is & Why It Matters Download PDF
T2 — Under the Hood: How It Works & How to Start Download PDF
T3 — The Homotopy Workflow & Tool Choice Download PDF
T4 — Project Organization: The Researcher as Project Manager Download PDF
T5 — Tracing, Mistakes & Managing Context Download PDF

Lectures 6 & 10 In preparation

These two decks are still being written. Each link below opens a placeholder; this page updates as the real decks are posted. The syllabus lists what they will cover.

Lecture 6 — Heterogeneous Agent Models via RL Placeholder
Lecture 10 — Case Studies Placeholder