• This is the 2021 version. For previous year' course materials, go to this branch
• Lecture and seminar materials for each week are in ./week* folders, see README.md for materials and instructions
• YSDA homework deadlines will be listed in Anytask (read more).
• Any technical issues, ideas, bugs in course materials, contribution ideas - add an issue
• Installing libraries and troubleshooting: this thread.

• week01 Word Embeddings

  • Lecture: Word embeddings. Distributional semantics. Count-based (pre-neural) methods. Word2Vec: learn vectors. GloVe: count, then learn. Evaluation: intrinsic vs extrinsic. Analysis and Interpretability. Interactive lecture materials and more.
  • Seminar: Playing with word and sentence embeddings
  • Homework: Embedding-based machine translation system

• week02 Text Classification

  • Lecture: Text classification: introduction and datasets. General framework: feature extractor + classifier. Classical approaches: Naive Bayes, MaxEnt (Logistic Regression), SVM. Neural Networks: General View, Convolutional Models, Recurrent Models. Practical Tips: Data Augmentation. Analysis and Interpretability. Interactive lecture materials and more.
  • Seminar: Text classification with convolutional NNs.
  • Homework: Statistical & neural text classification.

• week03 Language Modeling

  • Lecture: Language Modeling: what does it mean? Left-to-right framework. N-gram language models. Neural Language Models: General View, Recurrent Models, Convolutional Models. Evaluation. Practical Tips: Weight Tying. Analysis and Interpretability. Interactive lecture materials and more.
  • Seminar: Build a N-gram language model from scratch
  • Homework: Neural LMs & smoothing in count-based models.

• week04 Seq2seq and Attention

  • Lecture: Seq2seq Basics: Encoder-Decoder framework, Training, Simple Models, Inference (e.g., beam search). Attention: general, score functions, models. Transformer: self-attention, masked self-attention, multi-head attention; model architecture. Subword Segmentation (BPE). Analysis and Interpretability: functions of attention heads; probing for linguistic structure. Interactive lecture materials and more.
  • Seminar: Basic sequence to sequence model
  • Homework: Machine translation with attention

• week05 Transfer Learning

  • Lecture: What is Transfer Learning? Great idea 1: From Words to Words-in-Context (CoVe, ELMo). Great idea 2: From Replacing Embeddings to Replacing Models (GPT, BERT). (A Bit of) Adaptors. Analysis and Interpretability. Interactive lecture materials and more.

• week06 Domain Adaptation

  • Lecture: General theory. Instance weighting. Proxy-labels methods. Feature matching methods. Distillation-like methods.
  • Seminar+Homework: BERT-based NER domain adaptation

• week07 Model deployment, compression & acceleration

  • Lecture: how nlp models get deployed; how (and why) make your model faster and/or smaller
  • No assignment this time; instead, we showcase running a simple ML model in your browser

• week08 Large Language Models & Their Implications

  • Lecture: more BERTology; Large language models: GPT-3, OPT, BLOOM; in-context learning, prompt engineering, parameter-efficient fine-tuning
  • Practice: prompt engineering and LoRA on a 6.7B model in colab

More TBA

Course materials and teaching performed by

• Elena Voita - course admin, lectures, seminars, homeworks
• Boris Kovarsky - lectures, seminars, homeworks
• David Talbot - lectures, seminars, homeworks
• Sergey Gubanov - lectures, seminars, homeworks
• Just Heuristic - lectures, seminars, homeworks