Curriculum / DescriptionsCOMP_SCI 397: Seminar in Statistical Language Modeling
VIEW ALL COURSE TIMES AND SESSIONS
Prerequisites
CS-349 (or equivalent) and intermediate PythonDescription
Statistical language models assign probabilities to sequences of words, and are used in systems that perform text summarization, machine translation, question answering, information extraction, text generation (dialogue system) and many other tasks. In recent years, language models based on deep neural networks have dramatically improved the state of the art. This course will cover the fundamental technologies comprising statistical language models, such as word embedding methods, recurrent neural networks and transformers, and selected natural language processing tasks by reading canonical research papers that helped define the field. Students will be required to read and present research papers and to complete a substantial research project. This is a research-focused class, with several groups submitting papers to major conferences in the past.
TEXTBOOK: None (the class consists of reading/presenting research papers)
COMPUTER USAGE: Python, PyTorch, GPUs and Huggingface codebase
GRADING: A/B/C
COURSE INSTRUCTOR: David Demeter
COURSE COORDINATOR: David Demeter
