榫欐簮鏈熷垔缃?/p>
http://www.qikan.com.cn
鐢ㄦ渶浼樺寲鏂规硶瑙ｅ喅锛及绁炵粡缃戠粶璁粌闂
浣滆�咃細鏉?/p>
缈?/p>
鑻?/p>
鎴?/p>
鏉ユ簮锛氥�婄數鑴戠煡璇嗕笌鎶�鏈?/p>
路
瀛︽湳浜ゆ祦銆?/p>
2008
骞寸
12
鏈?/p>
鎽樿锛?/p>
BP
绁炵粡缃戠粶鍙互鏈夋晥鍦板闈炵嚎鎬х郴缁熻繘琛岄�艰繎锛屼絾鏄紶缁熺殑鏈�閫熶笅闄嶆悳绱㈡柟娉曞瓨
鍦ㄦ敹鏁涢�熷害鎱㈢殑闂銆傛湰鏂囨彁鍑烘妸
BP
绁炵粡缃戠粶杞寲涓烘渶浼樺寲闂锛岀敤涓�绉嶅叡杞搴︾畻娉曚唬鏇?/p>
鏈�閫熶笅闄嶆硶杩涜鎼滅储杩唬锛屾瀬澶у湴鎻愰珮浜嗘敹鏁涢�熷害銆?/p>
鍏抽敭璇嶏細绁炵粡缃戠粶锛?/p>
鏈�浼樺寲锛涗竴绉嶅叡杞搴︾畻娉?/p>
涓浘鍒嗙被鍙凤細
TP183
鏂囩尞鏍囪瘑鐮侊細
A
鏂囩珷缂栧彿锛?/p>
1009-3044(2008)12-20000-00
Training BP Neural Network using optimization methods
LI Xiang ,SU Cheng
(College of computer science,China University of Mining and Technology, Xuzhou
221000,China)
Abstract
锛?/p>
BP neural network can efficiently approximate any nonlinear system, but there is a
problem of inefficient learning speed with the conventional steepest descent algorithm. In this paper,
we try to convert neural network to an optimization model, and apply conjugate gradient algorithm to
it to bring a faster learning speed.
Keywords
锛?/p>
Neural network
锛?/p>
Optimization
锛?/p>
Conjugate gradient algorithm
1 BP
绁炵粡缃戠粶妯″瀷
BP(
鍓嶉寮?/p>
)
绁炵粡缃戠粶缁撴瀯绠�鍗?/p>
,
鍙搷浣滄�у己
,
鑳芥ā鎷熶换鎰忕殑闈炵嚎鎬ц緭鍏ヨ緭鍑虹郴缁燂紝鏄洰鍓嶅簲
鐢ㄥ箍娉涚殑绁炵粡缃戠粶妯″瀷銆?/p>
BP
缃戠粶鐢辫緭鍏ュ眰
i
銆侀殣鍚眰
j
銆佽緭鍑哄眰
k
鍙婂悇灞備箣闂寸殑鑺傜偣杩炴帴鏉冪粍
鎴愶紝绁炵粡鍏冩嫇鎵戝鍥?/p>
1
锛?/p>
缃戠粶鐨勫涔犺繃绋嬬敱淇℃伅姝ｅ悜浼犳挱鍜岃宸弽鍚戜紶鎾瀯鎴?/p>
: