Deep Learning for Natural Language Processing

4 credits

Deep learning methods for Natural Language Processing (NLP). Covers static word embeddings, feed-forward networks for text, recurrent neural networks, transformers, pre-training and transfer learning, with applications including sentiment analysis, translation, generation. Prerequisite: LING 572 or equivalent machine learning course; knowledge of programming in Python; experience with Linux/Unix commands; familiarity with linear algebra; and knowledge of multivariable calculus (especially partial derivatives/gradients of multivariable functions). Offered: Sp.