Natural Language Processing - Introduction (pdf)
Twenty courselets from n-gram models to transformers, decoding, fine-tuning and ethics, with notebooks and quizzes
|
Natural Language Processing - Introduction (pdf)
|
2 M |
|
Foundations and Statistical Language Models (pdf)
|
376.83 K |
|
Neural Networks - Introduction (pdf)
|
1.23 M |
|
NLP - Neural Language Models (pdf)
|
301.68 K |
|
NLP - Recurrent Neural Networks (pdf)
|
309.36 K |
|
NLP - Sequence-to-Sequence Models (pdf)
|
431.84 K |
|
NLP - The Transformer Revolution (pdf)
|
395.77 K |
|
The Transformer Architecture (pdf)
|
2.2 M | |
|
|
14.55 M |
|
NLP - Pre-trained Language Models (pdf)
|
361.71 K |
|
NLP - Advanced Transformers (pdf)
|
455.89 K |
|
NLP - Tokenization and Subword Models (pdf)
|
335.17 K |
|
NLP - Decoding Strategies (pdf)
|
466.13 K |
|
NLP Decoding Strategies (pdf)
|
2.45 M |
|
NLP Decoding Strategies (pdf)
|
1.25 M |
|
NLP - Fine-tuning and Prompt Engineering (pdf)
|
419.83 K |
|
NLP - Efficiency and Deployment (pdf)
|
517.48 K |
|
NLP - Ethics and Future Directions (pdf)
|
420.29 K |
|
NLP - Predicting the next word (pdf)
|
703.25 K |
Next-word prediction is the training task of every language model in the course, from bigram counts to transformers. Topics are n-gram models and smoothing, word embeddings, recurrent networks, sequence-to-sequence models, attention, pre-training with BERT and GPT, tokenization, decoding, fine-tuning, efficiency and ethics, with notebooks in Python for several of them. The 2026 edition of the next-word material is course 100100, The Next Word Problem - Full Course. The courselets in order: