A PDP Network For Learning UNIX Programs

Michael Frank
Psych 172 - Prof. Rumelhart
Stanford University
March 14, 1990

Abstract

A user-friendly program was created to connect a recurrent PDP network to arbitrary UNIX programs, in such a way that the network could learn through back-propagation to predict the program's textual response to characters in an input stream. In numerous experiments, the network was able to learn, for an impressive variety of simple programs and input regimens, to predict exactly what the next text character produced by the UNIX program would be. Results of these experiments are described, and a proposal is made for future work that would see if a PDP network could be made to explore complex UNIX programs on its own. The UNIX environment is promoted as a good testbed "world" that proposed mind-emulating programs could learn to explore.

Citation information

M. Frank. ``A PDP Network for Learning UNIX Programs.'' Term paper for David E. Rumelhart, March 1990.

@unpublished{Frank-90,
	author = {Michael Frank},
	title = {A {PDP} Network for Learning {UNIX} Programs},
	note = {Term paper for Psychology 162, David E. Rumelhart,
		Stanford University},
	month = mar,
	year = 1990
}

A PDP Network For Learning UNIX Programs

Abstract

Citation information

For hardcopy, email mpf@ai.mit.edu.