|
Workshop Program |
|||
|
The program for the workshop consists of a series of tutorial lectures on specific connectionist models followed by structured modeling sessions in the computer lab. The network architectures covered will include interactive-activation networks, backpropagation networks, simple-recurrent networks, Hopfield networks, self-organizing feature maps, and coupled-oscillator models. Lab sessions will consist of a series of exercises using the BrainWave simulator. The workshop coordinators are Dr. Devin McAuley (Department of Psychology, Ohio State University) and Dr. Simon Dennis (School of Psychology, University of Queensland). The Cognitive Science Center faculty sponsor is Professor Mari Jones (Department of Psychology, Ohio State University).
|
|||
|
|
|||
|
8:00 - 8:30 |
Registration |
||
|
8:30 - 10:00 |
Lecture |
Introduction to Connectionist Models: Tutorial lecture on the main concepts and themes in the connectionist-modeling field. The topics covered prepare the participant for the first set of laboratory exercises. |
|
|
10:00 -10:30 |
Break |
||
|
10:30 - 12:30 |
Laboratory |
Introduction to the BrainWave simulator: A step by step introduction to the BrainWave neural-network software. By the end of the lab, participants will be able to construct simple networks using the BrainWave package and perform model simulations. |
|
|
12:30 - 2:00 |
Lunch |
||
|
2:00 - 3:00 |
Lecture |
Interactive Activation models: Tutorial lecture on an interactive-activation (IA) model of letter perception. The model demonstrates the interaction of bottom-up and top-down information in perception. It has been used to account for a range of contextual aspects of letter perception. |
|
|
3:00 - 3:30 |
Break |
||
|
3:30 - 5:00 |
Laboratory |
Context effects in letter perception: A series of exercises exploring the properties of the IA model. Simulations include demonstrations of different aspects of human letter perception, including the word-superiority effect. |
|
|
6:00 - 7:30 |
Dinner |
||
|
7:30 - 9:00 |
Discussion |
Group formation: In the evening session, we will discuss a set of simple modeling problems and form small research groups. Each group will select a problem to work on during the project session on Sunday. |
|
|
|
|||
|
8:30 - 9:30 |
Lecture |
Backpropagation models: Introduction to supervised learning in neural networks and the backpropagation algorithm. Topics covered include linear separability, error minimization, and interference effects. |
|
|
9:30 - 10:00 |
Break |
||
|
10:00 - 12:00 |
Laboratory |
Learning by example: A series of exercises on learning in connectionist models. Backpropagation simulations contrast hard and simple classification problems, and demonstrate interference effects such as the Stroop phenomenon. |
|
|
12:00 - 1:00 |
Lunch |
||
|
1:00 - 3:00 |
Laboratory |
Projects: Participants apply their knowledge to their selected modeling problem in a structured setting. |
|
|
3:00 - 3:30 |
Break |
||
|
3:30 - 4:30 |
Lecture |
Model survey: The final lecture provides an overview to a range of architectures that have been instrumental in the development of the field. These include the Hopfield network, the simple-recurrent network, and the self-organizing map. |
|
|
|
|||
Saturday October 3rd
