A tentative plan for the schedule (topics, deadlines, etc.) is available here.
Schedule (subject to change as the term progresses)
- Resource
- PyTorch Tutorial (w/ links to Colab notebook)
- Jan 6
- Course Overview
- Eisenstein 1
- Jan 8
- Machine Learning Review - linear classification (notes)
- Eisenstein 1, J+M 4
- Jan 9
- Problem Set 0 due
- Jan 13
- Machine Learning Review - logistic regression, perceptron, SVM (notes)
- Eisenstein 2.0-2.5, 4.1, 4.3-4.5, J+M 5
- Jan 15
- Machine Learning Review - mutliclass classification
- Eisenstein 2.0-2.5, 4.1, 4.3-4.5, J+M 5
- Jan 17
- Project 0 due
- Jan 27
- Neural Networks - Feedforward, optimization (notes)
- Eisenstein 2.6, 3.1-3.3, J+M 7, Goldberg 1-4, J.G. Makin - Backpropagation
- Jan 29
- Word Embeddings
- Eisenstein 3.3.4, 14.5, 14.6, J+M 6, Goldberg 5
- Jan 30
- Problem Set 1 due
- Feb 3
- Sequence Models - HMM, Viterbi (notes)
- Eisenstein 7.0-7.4, J+M 17, J+M A
- Feb 5
- Conditional Random Fields (notes)
- Eisenstein 7.5, 8.3
- Feb 7
- Project 1 due
- Feb 10
- Recurrent Neural Networks (notes)
- Eisenstein 7.6, Goldberg 10-11, J+M 8
- Feb 12
- Guest Lecture - Multilingual Multi-Cultural LLMs - by Tarek Naous
- Feb 17
- Convolutional Neural Networks + Neural CRFs
- Eisenstein 3.4, 7.6, Goldberg 9
- Feb 19
- Encoder-Decoder + MT
- Eisenstein 18.3 - 18.5
- Feb 21
- Project 2 due
- Feb 24
- Attention+ Tokenization (notes)
- Wu+16 Google NMT, Holtzman+19 Degeneration
- Feb 26
- Transformer (notes)
- J+M 9, Vaswani+17 Transformers, Alammar’s blog post, Rush’s tutorial
- Mar 3
- Pretrained Language Models (part 1), Midterm Review
- J+M 11, ELMo BERT
- Mar 4
- Problem Set 2 due
- Mar 5
- Pretrained Language Models (part 2) + Ethics
- J+M 10, BART, GPT-3
- Mar 10
- Post-training of Language Models (part 3)
- InstructGPT
- Mar 12
- Open-source Language Models (part 4)
- Llama 3
- Mar 24
- Project Proposal due
- Mar 26
- Guest Lecture - Evaluation of LLM-generated Text - by Yao Dou
- Mar 28
- Project 3 due
- Apr 7
- In-class Midterm (practice exam)