Computer Science and Engineering 258A

Cognitive Modeling

Fall Quarter 2011

Tuesday/Thursday 2:00-3:30

CSE Building, Room 2154

Professor: Gary Cottrell

Office: CSE 4130

Phone: 858-534-6640

e-mail: gary@ucsd.edu

 

Course Description

 

CSE 258A is a graduate seminar devoted to recent research in Cognitive Modeling. This is not an introductory course, so the prerequisite is at least one graduate-level course (at UCSD or elsewhere) in some aspect of AI, Cognitive Science, Computational Cognitive Neuroscience, or closely related areas such as machine learning, statistics or pattern recognition.  Appropriate courses at UCSD include CSE 250A/B (Principles of AI: Learning), CSE 291 (Probabilistic Methods in AI and Machine Learning), Cognitive Science 260 (Pattern Recognition), BGGN 246 - Reinforcement Learning and Decision Making.  If you have no familiarity with at least one of neural nets, machine learning, or Bayesian modeling, you should probably not take this course.

This course will explore cognitive modeling with an emphasis on the student interests. It is intended for CSE, Cognitive Science and Interdisciplinary PhD in Cognitive Science Program students. The emphasis will be on models that attempt to match psychological data about the process in question. Depending on student interest, we will look at models pertaining to language processing, decision-making, memory, concept formation, sequential processes, and vision.

On the first day, we will have an organizational meeting and the instructor will lecture on cognitive modeling with reference to his own work. By the end of the first week, the students should each pick a paper to give a presentation on. Then, in each class meeting, one student will give a talk lasting about 45-60 minutes presenting a recent technical paper in detail.  In questions during the talk, and in the final 20 minutes, all seminar participants will critically discuss the paper and the issues raised by it.


The student should read at least two papers: A classic paper in the field that can be considered antecedent to the recent research paper (if none can be found, discuss this with the professor). Often it is necessary to read more than one antecedent paper. The more recent paper should be chosen from a high-quality conference (in this category, prize-winning papers are preferred!) or journal. If you choose a conference paper, since they are often only 6-8 pages, more reading beyond that paper may be required to understand what was done. Suggested papers are listed here.

In addition to presenting recent research, there will be a final project. The project guidelines and schedule, a lot of which was cribbed from Charles Elkan's project guidelines are here. Projects should be done in pairs or small groups. The more people involved, the more involved the project should be! The project is worth 50% of your grade. Here are some ideas for projects that I gave to the undergrads in CSE 190. Obviously, yours should be a bit more cutting-edge!!

The talk I gave in class is  here.

This is an example of the class schedule from a previous iteration of this course. It will be updated when I get your paper choices and dates.


DATE PRESENTER TITLE
PAPER
DISCUSSION
PAGE
SLIDES
September 27-Oct 08
Gary Cottrell
Introduction to some Cognitive Models

Discuss any paper here...










October 11
Yajaira Gonzales
The PDP approach to semantic cognition.
 
McClelland, J.L., & Rogers, T.T. (2003). The parallel distributed processing approach to semantic cognition. Nature Reviews Neuroscience, 4, 310-322. pdf

October 17
Gary Cottrell
Backprop tutorial


October 18
Vivek R
Deep dyslexia: A case study of connectionist neuropsychology
Plaut, D.C., & Shallice, T. (1993). Deep dyslexia: A case study of connectionist neuropsychology. Cognitive Neuropsychology, 10(5), 377-500. pdf

October 20
Akshay Balsubramani
Predicting Human Brain Activity Associated with the Meanings of Nouns.
Mitchell, T.M., Shinkareva, S.V., Carlson, A., Chang, K.-M., Malave, V.L., Mason, R.A., & Just, M.A. (2008). Predicting Human Brain Activity Associated with the Meanings of Nouns. Science, 320, 1191-1195. pdf (paper) pdf (supplement) website (supplement)

October 25
Anukool Junnarkar Reassessing working memory
MacDonald, M.C. & Christiansen, M.H. (2002). Reassessing working memory: A comment on Just & Carpenter (1992) and Waters & Caplan (1996)
pdf

October 27
Andrew Heiberg
Rational approximations to category learning.
Sanborn, A.N., Griffiths, T.L., & Navarro, D.J. (2010). Rational approximations to rational models: Alternative algorithms for category learning. Psychological Review, 117(4), 1144-1167. pdf

November  1
 Chris Fariss
Learning to learn causal models.
Charles Kemp, Noah D. Goodman, Joshua B. Tenenbaum (2010)  Learning to learn causal models.  Cognitive Science 34:1185–1243 pdf

November 3
Joonleng Tan
On the Nature and Scope of Featural Representations of Word Meaning
McRae, Ken and De Sa, Virginia 1997 On the nature and scope of featural representations of word meaning.  Journal of Experimental Psychology: General, 126(2), 99-130. pdf

November 8
Qihua Wu
How children learn to value numbers
Ramscar, M., Dye, M., Popick, H.M. & O’Donnell-McCarthy, F. (In press) How children learn to value numbers: Information structure and the acquisition of numerical understanding. PLoS ONE. pdf

November 10
Liam Kavanagh
Structural aspects of face recognition and the other-race effect.
O’Toole, A.J., Deffenbacher, K.A., Valentin, D., & Abdi, H. (1994). Structural aspects of face recognition and the other-race effect. Memory and Cognition, 22(2), 208-224. pdf

November 15
NO MEETING: Gary out of town.





November 17
Ryland Fallon

Understanding Normal and Impaired Word Reading: Computational Principles in Quasi-Regular Domains.
Plaut, D.C., McClelland, J.L., Seidenberg, M.S., & Patterson, K. (1996). Understanding Normal and Impaired Word Reading: Computational Principles in Quasi-Regular Domains. Psychological Review, 103, 56-105. pdf

November 22






November 24
Thanksgiving





November 29





December 1