Stat 360 course materials
Project title | Time | Location |
---|---|---|
Ryan Chen (TA) | Thursdays 3-4pm | Zoom: https://northwestern.zoom.us/j/3082623966 |
Professor Stadie | Posted on Canvas | Zoom: https://northwestern.zoom.us/j/3799597115 |
Lecture notes can be found on the course Canvas website.
Lecture | Date | Material | Readings |
---|---|---|---|
Week 1, Tuesday | January 7 | Introduction | Perplexity, Perplexity 2, Linear Models |
Week 1, Thursday | January 9 | Attention | Attention is All You Need, Attention Mechanisms, Attention with code, The Illustrated Transformer |
Week 2. Tuesday | January 14 | Transformers | Intro to Transformers, Discussion, Blog post, Skip connections, Layer normalization, Byte-Pair Encoding |
Week 2. Thursday | January 16 | Coding Transformers | |
Week 3, Tuesday | January 21 | BERT, GPT, LLAMA | Annotated GPT 2, GPT-2 paper, Adversarial attacks on GPT-2 LLAMA Paper, BERT, Understanding BERT |
Week 3, Thursday | January 23 | Prompt Tuning, chain of thought, hindsight chain of thought, backwards chain of thought, Graph of Thought, Tree of Thought, Training Chain-of-Thought via Latent-Variable Inference, prompt engineering | Language Models are Few Shot Learners, Zero shot chain of thought, LLMs are human-level prompt engineers, Tree of Thought, Chain of verification, Promptbreeder, Prompt Engineering Guide, Open-AI Prompting Guide |
Week 4, Tuesday | January 28 | Fine tuning, tool use, parameter-efficient fine tuning, LORA, Instruction Tuning (SFT), Neftune, quantization, Hugging face, fine tuning LLAMA | LORA, PEFT, Quantization, Quantization Blog, Alpaca, ToolEMU, NEFTune, Tool Use code, Model Calibration Mistral Fine Tuning, LLAMA fine tuning |
Week 4, Thursday | January 30 | ChatGPT and RLHF, rejection sampling, DPO, Gopher Cite | Stack LLAMA, Instruct GPT, PPO, DPO, RLHF References, TRL, Gopher Cite, Chain of hindsight |
Week 5, Tuesday | February 4 | No class | |
Week 5, Thursday | February 6 | RAG, when to use RAG vs SFT, lexacagraphical vs semantic search, sentence transformers, Retrieval transformers and long term memory in transformers, RAG Code. | RAG, RAG code, Sentence Transformers, Rag Evaluation, Advanced RAG |
Week 6, Tuesday | February 11 | Multi-Agent LLMs | Wonderful Team, ReWoo, Reasoning with Language Model, LLM+P, ReACT, AgentVerse |
Week 6, Thursday | February 13 | Prompt optimization, reflection, steering | |
Week 7, Tuesday | February 18 | GPTo1, LLMS + Tabular ML | |
Week 7, Thursday | February 20 | Vision Models, Asking questions about images. Conditional layer norm, FILM, CLIP, BLIP, LAVA | BLIP, CLIP, Llava, FiLM |
Week 8 Tuesday | February 25 | Stable Diffusion, tuning stable diffusion, Diffusion models, DDPM, classifier-free guidance | Stable Diffusion, Illustrated Stable Diffusion, Understanding Stable Diffusion Diffusion, DDPM, Diffusion as SDEs, Classifier Free Guidance, Diffusion code, More low level code |
Week 8 Thursday | February 27 | Frontiers, using LLMs to help diffusion models by planning out images. SD Edit, Diffusion in robotics. | |
Week 9, Tuesday | March 4 | Life is Worse with LLMs | |
Week 9, Thursday | March 6 | No class. Complete take-home final exam |
Project title | Date released | Due date |
---|---|---|
Assignment 1: Transformers | Jan 9 | Jan 21 |
Assignment 2: Prompt Tuning | Jan 22 | Jan 30 |
Assignment 3: SFT | Jan 31 | Feb 11 |
Assignment 4: RLHF | Feb 13 | Feb 24 |
Assignment 5: RAG | Feb 24 | March 4 |
Take-home final exam | March 5 | Due Friday, March 7th |