0
Àá½Ã¸¸ ±â´Ù·Á ÁÖ¼¼¿ä. ·ÎµùÁßÀÔ´Ï´Ù.

½Å°æ¸Á¸ðµ¨ÀÇ ÇнÀ ¾Ë°í¸®Áò¿¡ °üÇÑ ¿¬±¸

A Study on Learning Algorithm for Neural Network

Áö»ê°£È£º¸°ÇÀü¹®´ëÇÐ³í¹®Áý 1997³â 15±Ç 0È£ p.189 ~ 196
KMID : 0613219970150000189
Á¶ÀçÇö (  ) - Áö»êÀü¹®´ëÇÐ Àü»êÁ¤º¸Ã³¸®°ú

Abstract

Among the neural network, Hopfield network is recurrenct network that emboies a profound physical principle. The output of each neuron in the network is fed back to all other neuron but is no self-feedback.
Because of the efficience in the associative memory, the model can be applied to the problem of pattern recognition. In the paper, we use the hopfield network to recognize the noised pattern and we show that the proposed algorithm is fast in the recognition late.
KeyWords

¿ø¹® ¹× ¸µÅ©¾Æ¿ô Á¤º¸
µîÀçÀú³Î Á¤º¸