Computational Modeling

PSYC 575

Instructor
Michael Young
Contact: 453-3567, meyoung@siu.edu
Office hours: 2:00 - 3:30 Tu/Th, 9-12 W
Location: 271F LSII

Readings

Doherty, M.E., & Brehmer, B. (1997). The paramorphic representation of clinical judgment: A thirty-year retrospective. In W.M. Goldstein and R.M. Hogarth (Eds.), Research on Judgment and Decision Making: Currents, Connections, and Controversies, Cambridge, Cambridge University Press.
Elman, J. L. (1990). Finding structure in time. Cognitive Science, 14, 179-211.
French, R.M. (1999). Catastrophic forgetting in connectionist networks. Trends in Cognitive Sciences, 3, 128-135.
Glover, S. & Dixon, P. (2004). Likelihood ratios: A simple and flexible statistic for empirical psychologists. Psychonomic Bulletin & Review, 11, 791-806.
Goodman, S.N., & Royall, R. (1988). Evidence and scientific research. American Journal of Public Health, 78, 1568-1574.
Hill, T., O'Connor, M., & Remus, W. (1996). Neural networks models for time series forecasts. Management Science, 42, 1082-1092
Hinton, G.E., McClelland, J.L., & Rumelhart, D.E. (1986). Distributed Representations. In D.E. Rumelhart, J.L. McClelland, & The PDP Research Group, Parallel Distributed Processing: Volume 1 (pp. 77-109).
Jacobs, R. A., Jordan, M. I., Nowlan, S. J., & Hinton, G. E. (1991). Adaptive mixtures of local experts. Neural Computation, 3, 79-87.
Jäkel, F., Schölkopf, B., & Wichmann, F.A. (2008). Generalization and similarity in exemplar models of categorization: Insights from machine learning. Psychonomic Bulletin & Review, 15, 256-271.
Jordan, M.I. (1986). An introduction to linear algebra in parallel distributed processing. In D.E. Rumelhart, J.L. McClelland, & The PDP Research Group, Parallel Distributed Processing: Volume 1 (pp. 77-109).
Killeen, P.R. (2001). Modeling games in the 20th century. Behavioural Processes, 54, 33-52.
Pitt, M.A. & Myung, I. (2002). When a good fit can be bad. Trends in Cognitive Sciences , 6, 421-425.
Platt, J.R. (1964). Strong inference. Science, 146, 347-353.
Poggio, T., & Girosi, F. (1990). Regularization algorithms for learning that are equivalent to multilayer networks. Science, 247, 978-982.
Price, Spitznagel, et al. (2000). Applying artificial neural network models to clinical decision making. Psychological Assessment, 12, 40-51.
Sarle, W.S. (1994). Neural networks and statistical models. In Proceedings of the Nineteenth Annual SAS Users Group International Conference.
Shull, R. L. (1991). Mathematical description of operant behavior: An introduction. In I. H. Iversen & K. A. Lattal (Eds.), Experimental Analysis of Behavior (Vol. 2, pp. 243-282). New York: Elsevier.
Tassoni, C. J. (1995). The least mean squares network with information coding: A model of cue learning. Journal of Experimental Psychology: Learning, Memory and Cognition, 21, 193-204.
Tyree, E.W., & Long, J.A. (1998). A Monte Carlo evaluation of the moving method, k-means, and two self-organizing neural networks. Pattern Analysis and Applications, 1, 79-90.
West, P.M., Brockett, P.L., & Golden, L.L. (1997). A comparative analysis of neural networks and statistical methods for predicting consumer choice. Marketing Science, 16, 370-391.

Extra reading for those interested in a bit more on mixed-effects modeling:

Bagiella, E., Sloan, R.P., Heitjan, D.F. (2000). Mixed-effects models in psychophysiology. Psychophysiology, 37, 13-20. [A brief description of the relative strengths and weaknesses of repeated measures ANOVA, MANOVA, and linear mixed effects approaches to analyzing repeated measures data]
Cudeck, R., & Harring, J.R. (2007). Analysis of nonlinear patterns of change with random coefficient models. Annual Review of Psychology, 58, 615-637. [A brief introduction to nonlinear mixed effects modeling of psychological data]

Course Details
This course will survey various modeling techniques and their application to the field of psychology. Modeling is a method that strives to create mathematical descriptions of relationships between variables. Thus, this course will be of interest to anyone involved in psychology, artificial intelligence, economics, statistics, cognitive neuroscience, or one of the social sciences.

An integral part of the course will be the hands-on creation of models. You will use build small models and test their behavior. The final project in the class requires the development and presentation of a model (using any tool you wish to use) of a specific problem of interest to you. Teamwork will be allowed and encouraged.

There will be no exams for the course. Your final grade is based on your scores on the exercises (50%), class participation which may include pop-quizzes (20%), and your final project (30%).

Makeup/Late Policy, Complaints, and Cheating
In order to turn in something late without a penalty, you will need to provide a completed explanation of absence along with appropriate documentation (e.g. excuse signed by medical professional along with phone # and patient id, copy of funeral notice, police report). Apologies, but requiring documentation for all types of absences is the fairest policy. Late assignments without accompanying documentation accrue a late penalty.

Complaints and cheating will be handled in accordance with the policies outlined in the Student Code of Conduct.

Persons with disabilities
If you have a documented disability requiring special accommodations for assignments, contact me within the first two weeks of class so special arrangements can be made.

Dates	Topic	Reading
1/13 - 15	Course overview, Introduction to connectionism as a statistical technique Exercise 1 - Installing "JMP" (due 1/22)	Platt (1964) Sarle (1994)
1/20 - 22	Computational Basics I - Multiple linear regression, standard solutions Exercise 2 - Linear regression (due 1/29)	Doherty & Brehmer (1997)
1/27 - 29	Computational Basics II - Nonlinear model fitting with a single predictor and criterion variable Exercise 3 - Nonlinear fits (due 2/5)	Shull (1991) pp. 1-46 and 143-151 of Prism Book (575 home page)
2/3 - 5	Minimizing error versus Maximizing likelihood. Plus a brief introduction to issues in repeated measures fitting. Exercise 4 - Maximum likelihood estimation (due 2/12)	Glover & Dixon (2004) Goodman & Royall (1988)
2/10 - 17	Pattern Association I & II - Matrix multiplication and pattern association (multiple predictors and multiple criterion vars). Exercise 5 - Pattern association (due 2/19)	Jordan (1986) Tassoni (1995)
2/19	Models and representation	Hinton et al. (1986)
2/24 - 26	Backpropagation (using latent or hidden variables) and applications Exercise 6 - Backpropagation (due 3/19)	Price et al. (2000) French (1999) West et al. (1997)
3/3 - 5	Other computational modeling approaches Discussion of Final Project requirements	Killeen (2001)
Spring Break
3/17 - 19	Model selection and cross validation Exercise 7 - Cross validation (due 3/26)	Pitt & Myung (2002)
3/24 - 26	Temporal Learning I - incremental time series analysis	Hill et al. (1996)
3/31 - 4/2	Temporal Learning II and applications	Elman (1990)
4/7 - 9	Unsupervised Learning (incremental cluster analysis and PCA) Exercise 8 - Unsupervised learning (due 4/16)	Jacobs et al (1991) Sarle (1994) (revisited)
4/14 - 16	Hybrid networks (Nearest neighbor methods and Radial basis functions/kernel regression)	Poggio & Girosi (1990) Tyree & Long (1998)
4/21 - 23	RBF applications - Category learning	Jäkel (2008)
4/28 - 30 5/5 (12:50-2:50)	Presentation of Final Projects
5/6	Final project papers due

Exercises: There will be handouts describing what needs to be done. Each will be due the Thursday of the week after it was assigned. Final project will be due by 5/6 - grade will consist of presentation and paper. All exercises can be done with help from other students but each person will hand in their own UNIQUE writeup. The final project is an exception - group projects with a single multi-author paper are encouraged. Limit of 3 people in each group for the final project. Late Penalties: 10% for projects turned in on Friday, 20% for those turned in the following Monday, and a maximum penalty of 25% for those turned in later. Course material (syllabus, grades, helpful links) is available on-line at: http://www.psychology.siu.edu/bcs/young.html