|
|
CS 564 : Fall 1999 Brain Theory and Artificial Intelligence |
Tu Th: 11-12:20 am; OHE 100
Instructor:
Prof. Michael A. Arbib; HNB-03, (213) 740-9220, arbib@pollux.usc.edu.(Office hours: 11-12 Wednesdays, HNB 03.)
Teaching Assistants: Erhan Oztop, erhan@java.usc.edu, Salvador Marmol, smarmol@rana
This course provides a basic understanding of brain function, of the artificial neural networks which provide tools for a new paradigm for adaptive parallel computation, and of the Neural Simulation Language NSLJ which allows us to study biological and artificial neural networks in great detail. No background in neuroscience is required, nor is specific programming expertise, but knowledge of Java will enable students to extend the NSLJ functionality in interesting ways.
|
The Brain as a Network of Neurons |
|
Perceptual and Motor Schemas |
|
Neural Modeling in Perspective – Connectionism and Cognitive Science |
|
Didday Model of Winner-Take-All |
|
Introduction to NSL: modules; SCS schematic Capture System; The window interface and graphics |
|
Hopfield Networks, Constraint Satisfaction, and Optimization |
|
[NSLJ] Maxselector: a. How to run the model; b. How to write the model |
|
[NSLJ] Hopfield: a. How to run the model; b. How to write the model |
|
Adaptive Networks – Hebbian learning, Perceptrons; |
|
Adaptive Networks – Gradient Descent and Backpropagation |
|
[NSLJ] Backprop: a. How to run the model; b. How to write the model |
|
The Modeling Language NSLM; The Scripting Language NSLS |
|
Midterm |
|
Systems Concepts |
|
Feedback and the Spinal Cord; [NSLJ] |
|
The FARS model of Control of Reaching and Grasping 1 |
|
The FARS model of Control of Reaching and Grasping 2 |
|
Review of Midterm and NSL |
|
[NSLJ] Control of Saccades – Dominey 1 |
|
[NSLJ] Control of Saccades – Dominey 2 |
|
5pm-6:20pm – Make-up midterm |
|
Hedco Neurosciences Building: 10th Anniversary Celebration |
|
Visual Preprocessing; Lateral inhibition; Von Bekésy model |
|
Depth Perception; [NSLJ] Dev and House Models |
|
Self-Organizing Feature Maps; [NSLJ] Kohonen Maps |
|
Competition and Cooperation in Neural Nets |
|
Holiday: Thanksgiving Day |
|
Reinforcement Learning and Motor Control |
|
Robotic Learning |
|
[NSLJ] Cerebellar Adaptation of Movement Generation -- Prism Adaptation Model |
|
Brain Models on the Web (BMW) |
Texts:
A. Weitzenfeld, M.A. Arbib and A. Alexander, 1999-2000, NSL Neural Simulation Language, MIT Press (in press).
M.A. Arbib, Ed., 1995, The Handbook of Brain Theory and Neural Networks, MIT Press (paperback).
Supplementary reading: M.A. Arbib, 1989, The Metaphorical Brain 2: Neural Networks and Beyond, Wiley-Interscience.
One mid-term and a final will cover the entire contents of the readings as well as the lectures.
The final exam will cover all of the course, but emphasizing material not covered in the mid-term.
Distribution of Grades: NSL assignments and other homework or projects: 40%; Mid-term: 25%; Final Exam: 35%.