Error-correcting output coding (ECOC) is a method for converting a k-classsupervised learning problem into a large number L of two-class supervised learningproblems and then combining the results of these L evaluations. Previous researchhas shown that ECOC can dramatically improve the classi cation accuracy of supervisedlearning algorithms that learn to classify...
Artificial Intelligence (AI) planning techniques have been central to automating a gamut of tasks from the mundane route planning and beer production to the ethereal image processing of space-ship images. Of all the planning techniques, hierarchical- decomposition planning has been the technique most employed in industrial-strength planners. Hierarchical-decomposition planning is...
The well-known local adjustment algorithm for training a
threshold logic unit, TLU, is extended to a local adjustment
algorithm for training a network of TLUs Computer simulations
show that the extension is unsatisfactory.
A new logic for a committee of TLUs, called modified veto logic,
and a local adjustment algorithm...
A radix 2n non-restoring division algorithm is described. The
algorithm is designed to be compatible with hardware multiprecision
multiplication methods currectly used in high speed digital computers.
This enables the use of the same hardware, with only changes in
control logic, to be used to implement both multiplication and
division....
Specifications for a simulation language capable of modeling
fuzzy systems are presented. Actions occurring in the running
model are to be displayed on a graphics terminal in the .form of
modified E-nets. The concept of partial execution of predicates
is introduced in connection with the underlying continuously valued
logic. Fuzzy...
Tree systems are used in syntactic pattern recognition for
describing two-dimensional patterns. We extend results on tree
automata with the introduction of the subtree-invariant equivalence
relation R. R relates two trees when the appearance of one implies the
appearance of the other in similar trees. A new state minimizing
algorithm...
A new system called sequential/parallel matrix grammars
for two-dimensional pattern processing is introduced and studied.
Miscellaneous language operations such as union, catenation (row
and column), Kleene's closure (row and column) and substitutions
are investigated. The equivalence of sequential/parallel matrix
languages and finite-turn repetitive checking automata is established.
Hierarchies for both...
This research work investigates the use of production
systems as a model of parallel processing. The purpose of
the model is to provide a suitable medium within which parallel
processing systems can be systematically specified,
analyzed, and designed. Furthermore, the model provides a
suitable means for deriving implementations of synchronization...
The methodology of structured programming has
enabled rapid progress in many areas of theoretical
computer science. Structured programs are generally
easier to debug, test, prove and analyse. The development
of these achievements into commercially viable applications
and products has been slower than expected.
The primary reason is that most of...