Title :
A knowledge-based approach to supervised incremental learning
Author :
Fu, LiMin ; Hsu, Hui-Hunag ; Principe, Jose C.
Author_Institution :
Dept. of Comput. & Inf. Sci., Florida Univ., Gainesville, FL, USA
fDate :
27 Jun-2 Jul 1994
Abstract :
How to learn new knowledge without forgetting old knowledge is a key issue in designing an incremental-learning neural network. In this paper, we present a rule-based connectionist approach in which old knowledge is preserved by bounding weight modifications. In addition, some heuristics are developed for avoiding overtraining of the network and adding new hidden units. The feasibility of this approach is demonstrated for classification problems including the iris and the promoter domains
Keywords :
knowledge based systems; learning (artificial intelligence); neural nets; pattern classification; bounding weight modifications; classification; heuristics; incremental-learning neural network; knowledge-based system; rule-based connectionist; supervised incremental learning; Computer networks; Encoding; Iris; Learning systems; Multidimensional systems; Neural networks; Problem-solving; Real time systems; Uncertainty; Weight measurement;
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
DOI :
10.1109/ICNN.1994.374428
