±±¾©Ê¯ÓÍ»¯¹¤Ñ§Ôº2026ÄêÑо¿ÉúÕÐÉú½ÓÊÕµ÷¼Á¹«¸æ
²é¿´: 443  |  »Ø¸´: 3
±¾Ìû²úÉú 1 ¸ö LS-EPI £¬µã»÷ÕâÀï½øÐв鿴

baroshi

ľ³æ (ÕýʽдÊÖ)

[ÇóÖú] °ïæ²éһƪÎÄÕµÄSCIºÅ£¬Ð»Ð»¡£

Selective multiple kernel learning for classification with ensemble strategy

[ ·¢×ÔÊÖ»ú°æ http://muchong.com/3g ]
»Ø¸´´ËÂ¥

» ²ÂÄãϲ»¶

KeepFighting
ÒÑÔÄ   »Ø¸´´ËÂ¥   ¹Ø×¢TA ¸øTA·¢ÏûÏ¢ ËÍTAºì»¨ TAµÄ»ØÌû

muse

¾èÖú¹ó±ö (ÖªÃû×÷¼Ò)

¡¾´ð°¸¡¿Ó¦Öú»ØÌû

¸Ðл²ÎÓ룬ӦÖúÖ¸Êý +1
Selective multiple kernel learning for classification with ensemble strategy

×÷Õß:Tao Sun; Licheng Jiao; Fang Liu; Shuang Wang; Jie Feng

Pattern Recognition

¾í: 46  

ÆÚ: 11  

Ò³: 3081-90  

DOI: 10.1016/j.patcog.2013.04.003  

³ö°æÄê: Nov. 2013  

ÕªÒª

Multiple Kernel Learning (MKL) aims to seek a better result than single kernel learning by combining a compact set of sub-kernels. However, MKL with L1-norm easily discards the sub-kernels with complementary information and MKL with Lp-norm(pges2) often gets the redundant solution. To address these problems, a Selective Multiple Kernel Learning (SMKL) method, inspired by Ensemble Learning (EL), is proposed. Comparing MKL with Lp-norm(pges2), SMKL obtains a sparse solution by a pre-selection procedure. Comparing MKL with L1-norm, SMKL preserves the sub-kernels with complementary information by guaranteeing the high discrimination and large diversity of pre-selected sub-kernels. For quantifying the discrimination and diversity of sub-kernels, a new kernel evaluation is designed. SMKL reduces the scale of MKL optimization and saves the memory storing of the sub-kernels, which extends the scale of problem that MKL could solve. Specially, a fast SMKL method using Linfinity-norm constraint is focused, which needs no MKL optimization process. It means that the memory is hardly a limitation for MKL with the large scale problem. Experiments state that our method is effective for classification. [All rights reserved Elsevier].
×÷ÕßÐÅÏ¢

×÷ÕßµØÖ·: Tao Sun; Licheng Jiao; Shuang Wang; Jie Feng; Key Lab. of Intell. Perception & Image Understanding of Minist. of Educ. of China, Xidian Univ., Xi'an, China.

Fang Liu; Sch. of Comput. Sci. & Technol., Xidian Univ., Xi'an, China.
³ö°æÉÌ

Elsevier Science Ltd., UK
Àà±ð / ·ÖÀà

Ñо¿·½Ïò:Computer Science; Mathematics (ÓÉ Thomson Reuters Ìṩ)

·ÖÀà´úÂë:C1230L Learning in AI; C1180 Optimisation techniques

CODENTNRA8

ÊÜ¿ØË÷Òý:learning (artificial intelligence); optimisation; pattern classification

·ÇÊÜ¿ØË÷Òý:selective multiple kernel learning; ensemble strategy classificatio; single kernel learning; ensemble learning; EL; preselection procedure; Linfinity-norm constraint; SMKL method; MKL optimization process
ÎÄÏ×ÐÅÏ¢

ÎÄÏ×ÀàÐÍ:Journal Paper

ÓïÖÖ:English

Èë²ØºÅ:13704917

ISSN:0031-3203

²Î¿¼ÎÄÏ×Êý:32
ÆÚ¿¯ÐÅÏ¢

    Impact Factor (Ó°ÏìÒò×Ó): Journal Citation Reports®

ÆäËûÐÅÏ¢

´¦ÀíÀàÐÍ:Theoretical or Mathematical

ÎÄÏ׺Å:S0031-3203(13)00173-8
µã»÷ÎÒ»ØÌûÏÂÃæµÄ¡°ËÍTAºì»¨¡±£¬ËÍÎÒÒ»¶äºì»¨°É£¡
2Â¥2014-09-21 18:50:40
ÒÑÔÄ   »Ø¸´´ËÂ¥   ¹Ø×¢TA ¸øTA·¢ÏûÏ¢ ËÍTAºì»¨ TAµÄ»ØÌû

muse

¾èÖú¹ó±ö (ÖªÃû×÷¼Ò)

¡¾´ð°¸¡¿Ó¦Öú»ØÌû

¡ï ¡ï ¡ï ¡ï ¡ï
baroshi: ½ð±Ò+5, ¡ï¡ï¡ï¡ï¡ï×î¼Ñ´ð°¸ 2014-09-21 18:56:25
sunshan4379: LS-EPI+1, ¸ÐлӦÖú£¡ 2014-09-21 19:36:28
Selective multiple kernel learning for classification with ensemble strategy

×÷Õß:Sun, T (Sun, Tao)[ 1 ] ; Jiao, LC (Jiao, Licheng)[ 1 ] ; Liu, F (Liu, Fang)[ 2 ] ; Wang, S (Wang, Shuang)[ 1 ] ; Feng, J (Feng, Jie)[ 1 ]

PATTERN RECOGNITION

¾í: 46

ÆÚ: 11

Ò³: 3081-3090

DOI: 10.1016/j.patcog.2013.04.003

³ö°æÄê: NOV 2013

²é¿´ÆÚ¿¯ÐÅÏ¢
ÕªÒª

Multiple Kernel Learning (MKL) aims to seek a better result than single kernel learning by combining a compact set of sub-kernels. However, MKL. with L1-norm easily discards the sub-kernels with complementary information and MKL with Lp-norm(p >= 2) often gets the redundant solution. To address these problems, a Selective Multiple Kernel Learning (SMKL) method, inspired by Ensemble Learning (EL), is proposed. Comparing MKL with Lp-norm(p >= 2), SMKL obtains a sparse solution by a pre-selection procedure. Comparing MKL with Lp-norm, SMKL preserves the sub-kernels with complementary information by guaranteeing the high discrimination and large diversity of pre-selected sub-kernels. For quantifying the discrimination and diversity of sub-kernels, a new kernel evaluation is designed. SMKL reduces the scale of MKL optimization and saves the memory storing of the sub-kernels, which extends the scale of problem that MKL could solve. Specially, a fast SMKL method using L infinity-norm constraint is focused, which needs no MIC optimization process. It means that the memory is hardly a limitation for MKL with the large scale problem. Experiments state that our method is effective for classification. (C) 2013 Elsevier Ltd. All rights reserved.
¹Ø¼ü´Ê

×÷Õ߹ؼü´Ê:Ensemble learning; Kernel evaluation; Multiple kernel learning; Selective multiple kernel learning; Fast selective multiple kernel learning

KeyWords PlusIMENSIONALITY REDUCTION
×÷ÕßÐÅÏ¢

ͨѶ×÷ÕßµØÖ·: Feng, J (ͨѶ×÷Õß)
[ÏÔʾÔöÇ¿×éÖ¯ÐÅÏ¢µÄÃû³Æ]         Xidian Univ, Minist Educ China, Key Lab Intelligent Percept & Image Understanding, Xian 710071, Peoples R China.

µØÖ·:
[ÏÔʾÔöÇ¿×éÖ¯ÐÅÏ¢µÄÃû³Æ]         [ 1 ] Xidian Univ, Minist Educ China, Key Lab Intelligent Percept & Image Understanding, Xian 710071, Peoples R China
[ÏÔʾÔöÇ¿×éÖ¯ÐÅÏ¢µÄÃû³Æ]         [ 2 ] Xidian Univ, Sch Comp Sci & Technol, Xian 710071, Peoples R China

µç×ÓÓʼþµØÖ·:taosun@mail.xidian.edu.cn; lchjiao@mail.xidian.edu.cn; f63liu@163.com; shwang@mail.xidian.edu.cn; jiefeng0109@163.com
»ù½ð×ÊÖúÖÂл
»ù½ð×ÊÖú»ú¹¹        ÊÚȨºÅ
National Basic Research Program (973 Program) of China        
2013CB329402
National Natural Science Foundation of China        
61173092
61072106
61003198
Program for New Century Excellent Talents in University        
NCET-11-0692
Fundamental Research Funds for the Central Universities        
K50510020001
K50513100012
²é¿´»ù½ð×ÊÖúÐÅÏ¢   
³ö°æÉÌ

ELSEVIER SCI LTD, THE BOULEVARD, LANGFORD LANE, KIDLINGTON, OXFORD OX5 1GB, OXON, ENGLAND
Àà±ð / ·ÖÀà

Ñо¿·½Ïò:Computer Science; Engineering

Web of Science Àà±ð:Computer Science, Artificial Intelligence; Engineering, Electrical & Electronic
ÎÄÏ×ÐÅÏ¢

ÎÄÏ×ÀàÐÍ:Article

ÓïÖÖ:English

Èë²ØºÅ: WOS:000321232900017

ISSN: 0031-3203
ÆÚ¿¯ÐÅÏ¢

    Ŀ¼£º Current Contents Connect®

    Impact Factor (Ó°ÏìÒò×Ó): Journal Citation Reports®

ÆäËûÐÅÏ¢

IDS ºÅ: 175LT

Web of Science ºËÐĺϼ¯ÖÐµÄ "ÒýÓõIJο¼ÎÄÏ×": 32

Web of Science ºËÐĺϼ¯ÖÐµÄ "±»ÒýƵ´Î": 2
µã»÷ÎÒ»ØÌûÏÂÃæµÄ¡°ËÍTAºì»¨¡±£¬ËÍÎÒÒ»¶äºì»¨°É£¡
3Â¥2014-09-21 18:51:05
ÒÑÔÄ   »Ø¸´´ËÂ¥   ¹Ø×¢TA ¸øTA·¢ÏûÏ¢ ËÍTAºì»¨ TAµÄ»ØÌû

muse

¾èÖú¹ó±ö (ÖªÃû×÷¼Ò)

¡¾´ð°¸¡¿Ó¦Öú»ØÌû

Èë²ØºÅ: WOS:000321232900017
µã»÷ÎÒ»ØÌûÏÂÃæµÄ¡°ËÍTAºì»¨¡±£¬ËÍÎÒÒ»¶äºì»¨°É£¡
4Â¥2014-09-21 18:51:13
ÒÑÔÄ   »Ø¸´´ËÂ¥   ¹Ø×¢TA ¸øTA·¢ÏûÏ¢ ËÍTAºì»¨ TAµÄ»ØÌû
Ïà¹Ø°æ¿éÌø×ª ÎÒÒª¶©ÔÄÂ¥Ö÷ baroshi µÄÖ÷Ìâ¸üÐÂ
×î¾ßÈËÆøÈÈÌûÍÆ¼ö [²é¿´È«²¿] ×÷Õß »Ø/¿´ ×îºó·¢±í
[¿¼ÑÐ] 285Çóµ÷¼Á +4 AZMK 2026-03-27 6/300 2026-03-27 20:48 by w320357296
[¿¼ÑÐ] 277¹òÇóµ÷¼Á +5 1915668 2026-03-27 8/400 2026-03-27 19:53 by WYUMater
[¿¼ÑÐ] ²ÄÁÏÓ뻯¹¤£¨0856£©304ÇóBÇøµ÷¼Á +7 Çñgl 2026-03-27 7/350 2026-03-27 15:27 by JoeÂÊ
[¿¼ÑÐ] 274Çóµ÷¼Á +17 ¹Ë¾ÅóÏҪǫÐé 2026-03-24 23/1150 2026-03-27 15:16 by caszguilin
[¿¼ÑÐ] 0703Ò»Ö¾Ô¸9£¬³õÊԳɼ¨£º338£¬ËÄÁù¼¶Òѹý£¬ÓпÆÑо­Àú£¬Çóµ÷¼Á£¡ +3 Zuhui0306 2026-03-25 3/150 2026-03-27 14:09 by shangxh
[¿¼ÑÐ] 321Çóµ÷¼Á +4 è±Óñ~~ 2026-03-25 4/200 2026-03-27 11:41 by sanrepian
[¿¼ÑÐ] µ÷¼Á +3 Àî¼Îͼ¡¤S¡¤Â· 2026-03-27 3/150 2026-03-27 11:19 by wangjy2002
[¿¼ÑÐ] 0703»¯Ñ§338Çóµ÷¼Á£¡ +6 Zuhui0306 2026-03-26 7/350 2026-03-27 10:35 by shangxh
[¿¼ÑÐ] 317Çóµ÷¼Á +7 µ°»ÆÏÌÈâôÕ 2026-03-26 7/350 2026-03-27 02:29 by fmesaito
[¿¼ÑÐ] 071000ÉúÎïѧÇóµ÷¼Á£¬³õÊԳɼ¨343 +6 ССÌðÃæÍÅ 2026-03-25 6/300 2026-03-26 23:01 by ²»³Ôô~µÄ؈
[¿¼ÑÐ] Çóµ÷¼Á +5 «lty 2026-03-25 6/300 2026-03-26 20:49 by ²»³Ôô~µÄ؈
[¿¼ÑÐ] 324Çóµ÷¼Á +4 wysyjs25 2026-03-21 4/200 2026-03-26 20:38 by fmesaito
[¿¼ÑÐ] Ò»Ö¾Ô¸ÏÃÃÅ´óѧ»¯Ñ§Ñ§Ë¶307Çóµ÷¼Á +8 y7czhao 2026-03-26 8/400 2026-03-26 19:51 by ²»³Ôô~µÄ؈
[¿¼ÑÐ] »¯Ñ§¹¤³Ì085602 305·ÖÇóµ÷¼Á +17 RichLi_ 2026-03-25 17/850 2026-03-26 19:44 by plmuchong
[¿¼ÑÐ] 291 Çóµ÷¼Á +7 »¯¹¤2026½ì±ÏÒµÉ 2026-03-21 8/400 2026-03-26 11:25 by AlenQIN.
[¿¼ÑÐ] Ò»Ö¾Ô¸ÉϺ£½»´óÉúÎïÓëҽҩר˶324·Ö£¬Çóµ÷¼Á +6 jiajunX 2026-03-22 6/300 2026-03-25 23:05 by licg0208
[¿¼ÑÐ] ÉúÎïѧѧ˶Çóµ÷¼Á +7 СÑò˯×ÅÁË? 2026-03-23 10/500 2026-03-25 02:24 by Çå·ç·÷Ñï¡£ m
[¿¼ÑÐ] ²ÄÁÏר˶331Çóµ÷¼Á +4 Ïʵ±Å£ 2026-03-24 4/200 2026-03-24 15:58 by JourneyLucky
[¿¼ÑÐ] 333Çóµ÷¼Á +3 ALULU4408 2026-03-23 3/150 2026-03-23 19:04 by macy2011
[¿¼ÑÐ] Ò»Ö¾Ô¸¶«»ª´óѧ»¯Ñ§070300£¬Çóµ÷¼Á +7 2117205181 2026-03-21 8/400 2026-03-22 22:55 by chixmc
ÐÅÏ¢Ìáʾ
ÇëÌî´¦ÀíÒâ¼û