數理語言學:聯結論
Mathematical Linguistics: Connectionism


Fall 1998 Thursday 2:10-5:00 文學院215

TENTATIVE SYLLABUS (6/21/1998)

All about me:

James Myers (麥傑)
Office: 文學院247
Tel: 6471
Email: Lngmyers at ccu dot edu dot tw
WWW: http://www.ccunix.ccu.edu.tw/~lngmyers
Office hours: Tuesday 2pm-3pm, or by appointment.

OTHER WEB RESOURCES

Goals:

The ultimate focus of this class is connectionism in a solid mathematical and historical setting, but if you learn how to be more logical even when doing regular linguistics, the world will be a better place.

Readings: See end of syllabus.

Evaluation:

70% Homework
10% Participation
20% Final project

The homework comes in different flavors: sometimes questions about the readings, sometimes "math" exercises, sometimes computer exercises.

Participation means you talk. This is your class too.

The final project can be of two kinds, connectionist or non-connectionist. A connectionist project would consist of a working computer simulation of a connectionist model of some aspect of human language, based on the simulations that we use in the class. You'd hand in a disk with the simulation, and a paper explaining the empirical background and operation of the simulation. A non-connectionist project would be an analysis of some aspect of human language using formal language theory and/or learnability theory. You'd hand in a paper giving the empirical background and the formal analysis.

Schedule

("*" marks when homework is due)
("[]" mark optional readings)

Date Topics Readings
9/24 Introduction to math and mind
*10/1 Formal language theory Moll et al. ch. 1-2
*10/8 Applications to phonology, morphology and syntax Hammond;
Ellison;
Moll et al. section 8.2
*10/15 Learnability theory Morgan;
[Osherson et al.]
*10/22 Overview of connectionism Elman et al. ch. 2;
Pinker pp. 98-131
*10/29 The mathematics of connectionism Plunkett & Elman ch. 1-3;
[Anderson ch. 3, 5, 6]
*11/5 Backpropagation Rumelhart, Hinton & Williams;
Anderson pp. 275-277;
Plunkett & Elman ch. 4
11/12 ***NO CLASS***
*11/19 Autoassociation and generalization Plunkett & Elman ch. 5, 6
*11/26 Constraint satisfaction McClelland & Rumelhart ch. 3
*12/3 Rules without rules Plunkett & Elman ch. 11
*12/10 Syntax and learnability Elman (1994)
*12/17 Memory size and innateness Plunkett & Elman ch. 12
12/24 Fuzzy semantic categories Small (1997)
12/31 Constraint satisfaction and Optimality Theory Prince & Smolensky
1/7 PRESENTATIONS
1/14? FINAL PROJECTS DUE

LIST OF READINGS
[
Back to Top ]

Anderson, J. A. (1995) An Introduction to Neural Networks. MIT Press. [selections]

Davis, P. J. and Hersh, R. (1986) Descartes' Dream: The World According to Mathematics. Houghton Mifflin Co. [selections]

Ellison, T. M. (1995) "Phonological derivation in Optimality Theory." University of Edinburgh ms. Available through the Rutgers Optimality Archive.

Elman, J. L. (1994) "Implicit learning in neural networks: the importance of starting small." In Carlo Umilta and Morris Moscovitch [eds] Attention and Performance XV: Conscious and Nonconscious Information Processing, pp. 861-888. MIT Press.

Elman, J. L., Bates, E. A., Johnson, M. H., Karmiloff-Smith, A., Parisi, D. and Plunkett, K. (1996) Rethinking Innateness: A Connectionist Perspective on Development. MIT Press.

Hammond, M. (1993) "On the absence of category-changing prefixes in English." Linguistic Inquiry 24:562-567.

McClelland, J. L. and Rumelhart, D. E. (1989) Explorations in Parallel Distributed Processing: A Handbook of Models, Programs and Exercises. MIT Press.

Moll, R. N., Arbib, M. A., and Kfoury, A. J. (1988) An Introduction to Formal Language Theory. Springer-Verlag.

Morgan, J. L. (1986) From Simple Input to Complex Grammar. MIT Press. [selections]

Osherson, D. N., Stob, M., and Weinstein, S. (1989) "Learning theory and natural language." In Robert J. Matthews and William Demopoulos [eds] Learnability and Linguistic Theory. Kluwer.

Pinker, S. (1997) How the Mind Works. William Morrow. [selections]

Plunkett, K. and Elman, J. L. (1997) Exercises in Rethinking Innateness: A Handbook for Connectionist Simulations. MIT Press. [selections]

Prince, A. and Smolensky, P. (1997) "Optimality: from neural networks to universal grammar." Science 275:1604-1610.

Rumelhart, D. E., Hinton, G. E., and Williams, R. J. (1986) "Learning internal representations by error propagation." In Rumelhart and McClelland, 318-364.

Rumelhart, D. E. and McClelland, J. L. (1986) Parallel Distributed Processing. MIT Press. [selections]

Small, S. (1997) "Semantic category imprecision: a connectionist study of the boundaries of word meanings." Brain and Language 57(2):181-194.