king-menin/mipt-nlp2022

NLP course, MIPT

mipt-nlp2022

NLP course, MIPT

Course instructors

Anton Emelianov (login-const@mail.ru, @king_menin), Albina Akhmetgareeva (albina.akhmetgareeva@gmail.com)

Videos here

Exam questions here

Mark

$$final_mark=sum_i (max_score(HW_i)) / count(HWs) * 10, i==1..3$$

Lecture schedule

Week 1

• Lecture: Intro to NLP
• Practical: Text preprocessing,
• Video

Week 2

• Lecture: Word embeddings

Distributional semantics. Count-based (pre-neural) methods. Word2Vec: learn vectors. GloVe: count, then learn. N-gram (collocations) RusVectores. t-SNE.

• Practical: word2vec, fasttext
• HW1
• Video: lecture, seminar

Week 3

• Lecture: RNN + CNN, Text classification

Neural Language Models: Recurrent Models, Convolutional Models. Text classification (architectures)

• Practical: Classification with LSTM, CNN,
• Video

Week 4

• Lecture: Language modelling and NER

Task description, methods (Markov Model, RNNs), evaluation (perplexity), Sequence Labelling (NER, pos-tagging, chunking etc.) N-gram language models, HMM, MEMM, CRF

• Practical: NER,
• Video

Week 5

• Lecture: Machine translation, Seq2seq, Attention, Transformers

Basics: Encoder-Decoder framework, Inference (e.g., beam search), Eval (bleu). Attention: general, score functions, models. Bahdanau and Luong models. Transformer: self-attention, masked self-attention, multi-head attention.

• HW2, https://www.kaggle.com/c/mipt-nlp-hw2-2022

• Video

Week 6

• Lecture: Transfer learning in NLP

Bertology (BERT, GPT-s, t5, etc.), Subword Segmentation (BPE), Evaluation of big LMs.

• Practical: transformers models for classification task,
• Video

Week 7

Lecture & Practical: How to train big models? Distributed training

Training Multi-Billion Parameter Language Models. Model Parallelism. Data Parallelism.

• Practical: DDP example
• Video

Week 8

• Lecture: Question answering
• Practical: seminar QA , seminar chat-bots
• Video

Squads (one-hop, multi-hop), architectures, retrieval and search, chat-bots

Week 9

• Lecture: Summarization, simplification, paraphrasing
• Practical: summarization seminar
• HW3, https://www.kaggle.com/c/mipt-nlp-hw3-2022
• Video

Week 10

• Lecture: Multimodal NLP
• Video

Recommended Resources

En

• ruder.io
• Jurafsky & Martin
• Курс Лауры Каллмайер по МО для NLP
• Курс Нильса Раймерса по DL для NLP
• Курс в Оксфорде по DL для NLP
• Курс в Стенфорде по DL для NLP
• Reinforcment Learning for NLP

На русском (и про русский, в основном)

• Курс nlp в яндексе
• НКРЯ
• Открытый корпус
• Дистрибутивные семантические модели для русского языка
• Морфология
• Синтаксис
• Томита-парсер
• mathlingvo
• nlpub
• Text Visualisation browser

Literature

• Manning, Christopher D., and Hinrich Schütze. Foundations of statistical natural language processing. Vol. 999. Cambridge: MIT press, 1999.
• Martin, James H., and Daniel Jurafsky. "Speech and language processing." International Edition 710 (2000): 25.
• Cohen, Shay. "Bayesian analysis in natural language processing." Synthesis Lectures on Human Language Technologies 9, no. 2 (2016): 1-274.
• Goldberg, Yoav. "Neural Network Methods for Natural Language Processing." Synthesis Lectures on Human Language Technologies 10, no. 1 (2017): 1-309.