Machine Learning
Research in machine learning is very active at UNM. The computational
characterizations of learning assume many forms, genetic and emergent
models, data mining analyses, connectionist or neural network
techniques, and the more traditional symbol based Artificial
Intelligence approaches. In this later group there are three active
research efforts: analogical discovery, work done by Bill
Stubblefield and supervised by George Luger, diagnostic reasoning
and abduction, ongoing research by Carl Stern in collaboration with
George Luger, and information extraction research conducted by Steve
Verzi and supervised by Greg Heillman of the EECE Department. This later
work would also fit into the data mining and connectionist approaches to
learning.
[Back] [Home]