Course Overview
Course Description
Content
Objectives and Structure
Some generally useful materials
Assignments
Papers
Presentations
Students will participate in the presentation and discussion of the readings.Course Policies & Grading
Grading
Probability and information theory for linguists
Reading:
- Probability for linguists. John Goldsmith.
- Probabilistic models of grammar: phonology as information maximization. John Goldsmith.
- An Examination of Information Theory Yehoshua Bar-Hillel. Philosophy of Science, Vol. 22, No. 2 (Apr., 1955), pp. 86-105.
Prediction and entropy of printed English. Claude Shannon. Excerpts from Silicon Dreams. Robert Lucky.
Krenn and Samuelsson's manuscript, The Linguist's Guide to Statistics: Don't Panic is, despite the name, not for the faint of heart. If you feel mathematically comfortable and at ease, then it's a fine resource.
Finite state automata
Good webpage on FSAs.
Bayesian modeling and Minimum Description Length
Some other interesting links: Mike Dowman's website & papers
- MDL approach to grammar inference. Grunwald.
- Towards a new empiricism for linguistics. John Goldsmith.
Gold and PAC learning
Goldsmith and Riggle
Goldsmith-Xanthos-Baker
Three models for learning phonological categories. John Goldsmith and Aris Xanthos. Technical report TR 2008-08, Department of Computer Science. Submitted for review.
- Selections from
Methods in Structural Linguistics, by Zellig Harris John Goldsmith on Zellig Harris Fernando Pereira on Zellig Harris
Colin Wilson and Bruce Hayes
