DocumentCode :
1578943
Title :
Randomized learning in Y 𝔏-neural nets
Author :
Terekhoff, Serge A.
Author_Institution :
Russian Inst. of Tech. Phys., Russia
fYear :
1992
Firstpage :
457
Abstract :
The author proposes the perceptive artificial neural net, which consists of two different types of formal neurons with greatest simplicity. It is found that the net of the two simplest units (Y- and λ-neurons) is capable of representing some classes of net topologies and can resolve the general categorization problem. It is also shown how to implement well-known statistical physics methods to learn the Yλ-net, and some properties of learning trajectories are discussed. The simulated annealing technique is used at the learning session, and some advantages and disadvantages of this general method are pointed out
Keywords :
learning systems; neural nets; simulated annealing; Y L-neural nets; Y lambda -net; Y-neurons; lambda -neurons; learning trajectories; net topologies; perceptive artificial neural net; randomized learning; simulated annealing technique; statistical physics; Artificial intelligence; Artificial neural networks; Microscopy; Multilayer perceptrons; Neurons; Performance evaluation; Physics; Simulated annealing; Testing; Topology;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neuroinformatics and Neurocomputers, 1992., RNNS/IEEE Symposium on
Conference_Location :
Rostov-on-Don
Print_ISBN :
0-7803-0809-3
Type :
conf
DOI :
10.1109/RNNS.1992.268590
Filename :
268590
Link To Document :
بازگشت