Deep Learning for Natural Language Processing: A Gentle Introduction
Latest draft: August 17, 2022
The code discussed in the implementation chapters is available here.
- 08/17/2022: Added Appendix A. Many small changes.
- 07/28/2022: Added chapter 15 - implementing encoder-decoder methods. Many small changes throughout.
- 07/20/2022: Fixed typos and grammar in Chapter 13.
- 07/16/2022: Added chapter 13. Added historical background, references, and summaries to chapters 2 – 11.
- 07/12/2022: Fixed many typos throughout the book.
- 07/09/2022: Added summaries to all completed chapters.
- 07/07/2022: Four new chapters: Chapter 9 - text classification in PyTorch, Chapter 11 - POS tagging using RNNs in PyTorch, Chapter 14 - encoder-decoder methods, and Chapter 16 - neural architectures for NLP applications. Revised intro to match the new chapters.
- 03/09/2022: Fixed typos and grammar issues in Chapter 7.
- 03/07/2022: Fixed typos in Chapter 12. Added Chapter 7 with code for FFNNs.
- 02/18/2022: Completed Chapter 12; added the forward and Viterbi algorithms.
- 01/21/2022: Fixed several typos in Chapters 4 and 12. Started the CRF section.
- 01/14/2022: Added Chapter 4 (implementing LR from scratch and in PyTorch) and content to Chapter 12 on RNNs and LSTMs.
- 10/22/2021: Added Chapter 10 on transformer networks
- 07/06/2021: Completed Chapter 6 (best practices); added Appendix B (character encodings)
- 04/16/2021: Added Chapter 6.3: Activation Functions
- 04/13/2021: Added discussion of mini-batching in Chapter 6
- 03/25/2021: Fixed several typos in Chapter 5
- 03/19/2021: Added first draft of Chapter 5: Feed Forward Neural Networks
- 02/17/2021: Improved discussion in Section 3.7
- 02/05/2021: Added Section 3.6: Evaluation Measures for Multiclass Text Classification
- 01/27/2021: Added Chapter 7: Distributional Hypothesis and Representation Learning
You are welcome to use the slides any way you wish. But I would appreciate providing credit. And, optionally, dropping me a note if you found them useful.