This course presents an introduction to Natural language processing (NLP) with an emphasis on computational semantics i.e. the process of constructing and reasoning with meaning representations of natural language text.
The objective of the course is to learn about various topics in computational semantics and its importance in natural language processing methodology and research. Exercises and the project will be key parts of the course so the students will be able to gain hands-on experience with state-of-the-art techniques in the field.
The final assessment will be a combination of a group paper presentation (10%), two graded exercises (40%) and the project (50%). There will be no written exams.
Lectures: Fri 14:00-16:00 (the class will be in person)
Discussion Sections: Fri 16:00-17:00
Textbooks: We will not follow any particular textbook. We will draw material from a number of research papers and classes taught around the world. However, the following textbooks would be useful:
18.02 Class website is online!
| Lecture | Date | Description | Optional Readings | Events |
| 1 | 25.02 | Introduction | Diagnostic Quiz Answers to quiz | Presentation preference indication |
| 2 | 04.03 | The Distributional Hypothesis and Word Vectors | 1. Glove | |
| Voluntary | 04.03 | Matrix Calculus and Backpropagation by TAs | 1. CS231n notes on network architectures 2. CS231n notes on backprop 3. Learning Representations by Backpropagating Errors 4. Derivatives, Backpropagation, and Vectorization 5. Yes you should understand backprop |
|
| 3 | 11.03 | Word Vectors 2, Word Senses and Sentence Vectors (Recursive Neural Networks) |
1. Unsupervised Word Sense Disambiguation Rivaling Supervised Methods 2. Improving Vector Space Word Representations Using Multilingual Correlation |
|
| Voluntary | 11.03 | Python, PyTorch review session by TAs | 1. Review of Differential Calculus 1. Natural Language Processing (Almost) from Scratch |
|
| 4 | 18.03 | From words to sentences… Recurrent Neural Networks for Language Case Study: Language Modelling |
1. Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation | Assignment 1 released |
| 5 | 25.03 | NLU beyond a sentence Seq2Seq and Attention Case Study: Sentence Similarity, Textual Entailment and Machine Comprehension |
1. Massive Exploration of Neural Machine Translation Architectures 2. Bidirectional Attention Flow for Machine Comprehension |
|
| 6 | 01.04 | Syntax and Predicate Argument Structures (Semantic Role Labelling, Frame Semantics, etc.) |
1. Stanford’s Graph-based Neural Dependency Parser at the CoNLL 2017 Shared Task 2. Grammar as a foreign language |
|
| 7 | 08.04 | Predicate Argument Structures II (Semantic Role Labelling, Frame Semantics, etc.) |
Voluntary 1.Jointly Predicting Predicates and Arguments in Neural Semantic Role Labeling 2.Frame-Semantic Parsing |
|
| 08.04 | Discussion on Final Projects | 1. Practical Methodology (Deep Learning book chapter) | ||
| Easter | 15.04 | |||
| Easter | 22.04 | Project proposal due | ||
| 8 | 29.04 | Formal Representations of Language Meaning | 1.Compositional semantic parsing on semi-structured tables 2.Supertagging With LSTMs |
Assignment 1 due Assignment 2 release |
| 9 | 06.05 | Transformers and Contextual Word Representations (BERT, etc.) Guest lecture by Manzil Zaheer (Google) |
1. Big Bird: Transformers for Longer Sequences (Only cover the idea of sparse attention: don’t need to cover turing completeness and the theoretical results)) 2. BERT rediscovers the classical NLP pipeline |
|
| 10 | 13.05 | Question Answering | Voluntary 1. Reading Wikipedia to Answer Open-Domain Questions 2. Latent Retrieval for Weakly Supervised Open Domain Question Answering |
|
| 11 | 20.05 | Natural Language Generation Case Study: Summarization and Conversation Modelling |
1. Language Models are Unsupervised Multitask Learners 2. BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension |
Mid-term report due |
| 12 | 27.05 | Modelling and tracking entities: NER, coreference and information extraction (entity and relation extraction) | 1. End-to-end Neural Coreference Resolution 2. Improving Coreference Resolution by Learning Entity-Level Distributed Representations |
|
| 13 | 03.06 | Language + {Knowledge, Vision, Action} | 1. Knowledge Graph Embedding: A Survey of Approaches and Applications 2. Knowledge Enhanced Contextual Word Representations |
|
| 10.06 | Assignment 2 due | |||
| 15.07 | Project report due | |||
| 22.07 2-4pm |
Schedule | Poster session (gather town link) |
You can ask questions on moodle. Please post questions there, so others can see them and share in the discussion. If you have questions which are not of general interest, please don’t hesitate to contact us directly.
| Lecturer | Mrinmaya Sachan |
| Teaching Assistants | Alessandro Stolfo, Shehzaad Dhuliawala, Yifan Hou, Tianyu Liu |