SHORT DOCUMENT FOR USING IAC

Short Document for Using IAC

What is IAC?
About IACG.
Getting Start with CL.
Commands Useful for the Novice.
Sample Session.
Reference.

What is IAC:

IAC, stands for Interactive Activation and Competitive Learning, is a command-driven system designed by McClelland and Rumelhart for training and testing neural networks using interactive activation and competitive learning topology. IAC neural network is an extension of competitive learning paradigm and has been used in a number of applications, such as speech perception, word recognition, visual perception, and cell formation.

About IACG:

Preparing data files to run IAC is a complicated and time consuming process. To simplify the task, a program generator called IACG is developed to generate all files needed to run IAC. However, IACG is designed for solving cell formation problem only. The machine/part matrix, which is needed for cell formation, must be stored in a file. The format of this data file must be organized as follows:

R (number of rows - machines)
C (number of columns - parts)
Machine/Part Matrix (R x C binary matrix)

An sample file "Exa" is stored in the server for reference, where the machine/part matrix contains 12 machines and 10 parts.

Getting Start with IAC:

The "IAC" program and its utility "IACG" were stored in my course server under the directory "m528\systems\iac". First, create a directory (e.g., IAC) in your hard drive and then download all files into that directory. To access the programs, you must first quit the Window environment to DOS environment. Change your default directory to where IAC was stored. You must run "IACG" to generate the necessary files for use by IAC and then run IAC by typing "IAC file.tem file.str", where "file" is the corresponding file name you designed in the IAC run.

Commands Useful for the Novice:

IAC is a command-driven system; that is you must enter proper command to trigger the system. The following commands are useful for the novice:

input: to allow user to specify external inputs to units by name or number.

cycle: to run ncycles of processing.

test: to present a pattern from the pattern list for testing.

clear: to clear the display area.

reset: to reset activations of units in the network to resting level.

set/: to examine the current value of a variable and to set it to a new value if desired.

run: to pass a command to the command interpreter under which you are running the simulation program.

quit: to exit the program.

Sample Session:

The following sequence of commands (in bold face) can be used to select an unit for activating (input), specify input strength, and train the network (cycle):

reset (reset all activation values to 0)
input (specify an input pattern to present)
Do you want to reset all inputs? (y or n) y (answer yes)
Give unit name or number: M2 (M should be in capital case)
Enter input strength of M2: 1
Give unit name or number: Press "Enter" key
cycle (repeat until the weights are stabilized)

The above key strokes need to be repeated until all parts and machines have been classified.

References:

McClelland, J. L. and Rumelhart, D. E., Explorations in Parallel Distributed Processing: A Handbook of Models, Programs, and Exercises, The MIT Press, MA., 1988.