15-386/686 Neural Computation

Carnegie Mellon University

Spring 2010 School of Computer Science

Course Description

Neural Computation is an area of interdisciplinary study that seeks to understand how the brain computes to achieve natural intelligence. It seeks to understand the computational principles and mechanisms of intelligent behaviors and mental abilities -- such as perception, language, motor control, and learning -- by building artificial systems and computational models with the same capabilities. This course explores how neurons encode and process information, adapt and learn, communicate and compute at the individual level as well as at the levels of networks and systems. It will cover basic concepts in computational and neuronal modeling, information theory and coding, neural networks and learning, neural data analysis and decoding, signal processing and system analysis, statistical learning and probabilistic inference. Concrete examples will be drawn from the visual system and the motor system, including brain computer interface and neural decoding. Students will learn to perform quantitative analysis and perform computational experiments using Matlab. No prior background in biology is assumed. Prerequisites: 15-100, 21-120 or permission of instructor. 21-241 preferred but not required.

Course Information

Instructors	Office (Office hours)	Email (Phone)
Tai Sing Lee (Professor)	Mellon Inst. Rm 115	tai@cnbc.cmu.edu (412-268-1060)
	GHC 8017 (Tu/Thur 4:30-5:30 pm)
Anoopum S Gupta (TA)	GHC 7707 (Mon 3:30-5:00 pm)	anoopum@cmu.edu (919-244-4977)
Kai-min Kevin Chang (TA)	Baker 327B (Wed 3:30-5:00 pm)	kaimin.chang@gmail.com (412-268-3414)

Class location and time: GHC 4307. Tuesday, Thursday 3:00 p.m - 4:20 p.m.
Website: http://www.cnbc.cm u.edu/~tai/nc10.html (course info)
Course directory: /afs/andrew.cmu.edu/scs/cs/15-386 (homework and solutions)
Blackboard: http://www.cmu.edu/blackboard/ (Both 386/686 students should use 386 BB for general communications. 686 students should use 686 BB for additional works/communications desginated for 686.)

Handouts (including soft-copy of chapters of the following books).
Trappenberg T.P. (TTP) Fundamentals of computational neuroscience, 2nd edition, Oxford University Press (recommended).
Wilson H.R. (W) Spikes, decision and actions., Oxford University press 2004 (reference).
Dayan P, Abbott L (DA) Theoretical Neuroscience, MIT press, 2001. (reference, some chapters, more advanced).
Hertz J, Krogh A, Palmer RG (HKP) Introduction to the theory of neural computation., Addison Wesley 1991 (reference).

Grading Scheme

Evaluation	% of Grade
6 Assignments	60
Midterm	10
Term Project or Term Paper	30
Term Paper II for 686	15

Assignment

You will have 1 to 2 weeks to do each homework assignment. Homework report should be type-written.
Each homework assignment will have two parts. 686 students have to do both parts individually. 386 students are required to do one part only (individually). However, 386 students may work in pairs. Each team then need to complete both parts, and the grade will be equally divided among the partners. If a 386 student completes both parts individually, the higher score part will count.

Late Policy

Each student has ONE chance to turn in one of the six homework assignments late within one week without penalty.
No extension however is possible for the term project and term paper.

Term project/Term paper

Each 686 student has to do a term project individually, no partner allowed.
Term paper II is required of 686 students. It involves an in-depth study of a particular topic and an oral presentation of that topic in the form a 20 minute lecture. 686 students may work with one partner on Term Paper II.
Each 386 student can do a term project with one partner (or individually). Alternatively, the student can elect to write a survey paper individually.
Term project: The term project can be an implementation of a student's own idea, a model presented in a paper, or a new computational study that analyzes neural data, solves an applicational problem or addresses a scientific question. The Term Project will include 1) a proposal, 2) a progress report, and 3) a final report. The proposal should make clear the scientific question to be answered and provide a brief review of background literature, a description of the project goal and plans of actions. The proposal is due roughly around mid March, after Spring break. The final report should describe and discuss the project in a specific format provided (4-6 pages). Commented Matlab codes and additional output should also be submitted as supplementary materials in a different pdf/doc file and/or matlab files.
Term Project will be graded by the whole class.
Term Paper: The term paper (about 6-10 pages long) should be the outcome of an in-depth and extensive literature study of a particular topic, and should be presented in a 20-30 min lecture style oral presentation on that topic to the instructors and/or the class.
Note a 386 student can choose do term paper in lieu of a project, but a 686 student is required to do both.

Examinations

Midterm Quiz will test the materials covered in the class, and in a few specified required readings.
We plan not to have a Final Examination.

Syllabus

Date	Lecture Topic	Assignments
	NEURONS AND NEURAL SYSTEM ANALYSIS
T 1/12	1. Introduction
R 1/14	2. Neurons
T 1/19	3. Basic models of neurons
R 1/21	4. Synapses and Poisson models	HW 1 out
T 1/26	5. Neuronal tunings
R 1/28	6. Linear system analysis
T 2/2	7. Receptive field models and uses	HW 1 due. HW 2 out.
R 2/4	8. Network Analysis
	LEARNING IN NEURAL NETWORKS
T 2/9	9. Hebbian learning and PCA
R 2/11	10. Redundancy Reduction and Efficient Codes	HW 2 due. HW 3 out
T 2/16	11. Unsupervised learning
R 2/18	12. Supervised learning and classification
T 2/23	13. Reward and Reinforcement Learning	HW 3 due. HW 4 out.
	DECODING AND BRAIN COMPUTER INTERFACE
R 2/25	14. Hippocampus and memory
T 3/2	15. Spatial and trajectory coding
R 3/4	Midterm exam: Quiz 1	HW 4 due. HW 5 out.
M 3/8	Midterm Grade due 6 p.m.
T 3/9	Spring break
R 3/11	Spring break
T 3/16	16. Visual system and perception
R 3/18	17. Decoding visual system
T 3/23	18. Cognitive neural architecture
R 3/25	19. Decoding networks and semantics	HW 5 due. HW 6 out
T 3/30	20. Motor system and action
R 4/1	21. Neural prosthetics
	STATISTICAL LEARNING AND INFERENCE
T 4/6	22. Bayesian decision theory	HW 6 due.
R 4/8	23. Hierarhical models
T 4/13	24. Deep Belief Nets
R 4/15	Spring Carnival
T 4/20	25. Markov Random Fields
R 4/22	26. Hierarchical Inference
T 4/27	27. Cognitive structural learning
R 4/29	28. Review
M 5/3	Term Project and Term Paper deadline
X 5/X	Final Exam Day: Poster Presentation
R 5/13	Final Grade due.

Questions or comments: contact Tai Sing Lee or Kai-min Kevin Chang
Last modified on Thu Jan 14 12:45:58 EST 2010 by Kai-min