快速多视角特权协同随机向量函数连接网络

doi:10.3778/j.issn.1673-9418.2101075

计算机科学与探索 ›› 2022, Vol. 16 ›› Issue (10): 2320-2329.DOI: 10.3778/j.issn.1673-9418.2101075

快速多视角特权协同随机向量函数连接网络

吴天宇⁺(), 王士同

江南大学人工智能与计算机学院,江苏无锡 214122

收稿日期:2021-01-21 修回日期:2021-03-16 出版日期:2022-10-01 发布日期:2021-03-23
通讯作者: + E-mail: 6191610014@jiangnan.edu.cn
作者简介:吴天宇（1995—）,男,甘肃酒泉人,硕士研究生,主要研究方向为人工智能、模式识别。
王士同（1964—）,男,江苏扬州人,教授,博士生导师,CCF会员,主要研究方向为人工智能、模式识别等。
基金资助:
国家自然科学基金(61972181)

Fast Multi-view Privileged Random Vector Function Link Network

WU Tianyu⁺(), WANG Shitong

School of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi, Jiangsu 214122, China

Received:2021-01-21 Revised:2021-03-16 Online:2022-10-01 Published:2021-03-23
About author:WU Tianyu, born in 1995, M.S. candidate. His research interests include artificial intelligence and pattern recognition.
WANG Shitong, born in 1964, professor, Ph.D. supervisor, member of CCF. His research interests include artificial intelligence, pattern recognition, etc.
Supported by:
National Natural Science Foundation of China(61972181)

摘要/Abstract

摘要：

现实情况中通常会针对同一对象从不同途径或层面获得特征数据,称这样获得的数据为多视角数据。对于多视角数据的挖掘利用具有研究价值,与传统的单视角学习相比表现出一定优势。多视角学习（MVL）中一个重要的问题是如何在满足视角间互补情况下同时保持视角之间一致性。为解决上述问题,基于多视角学习和特权信息学习（LUPI）概念,以随机向量函数连接网络（RVFL）为基础,提出了一种快速多视角特权协同随机向量函数连接网络（FMPRVFL）来有效地解决多视角分类任务。该方法的基本思想是在平均情况下相互利用冗余视角的附加信息作为特权信息监督当前视角的分类。在此基础上设计的FMPRVFL的目标函数可以利用解析解对目标函数进行优化,从而使FMPRVFL训练速度更快。理论分析表明,与经典的多视角学习方法相比,FMPRVFL可以提供额外的泛化能力。在64个数据集上的实验结果表明,FMPRVFL在平均测试精度和运行时间上都优于比较方法。

关键词: 多视角学习（MVL）, 特权信息, 随机向量函数连接网络（RVFL）

Abstract:

In reality, feature data are usually obtained from different ways or levels for the same object, and the data obtained are called multi-view data. It is of research value to mine and utilize multi-view data and shows some advan-tages over traditional single-view learning. An important issue in multi-view learning (MVL) is how to meet the consistency between perspectives while maintaining complementarities between perspectives. In order to solve the above problem, based on the concepts including multi-view leaning and learning using privileged information (LUPI), a fast multi-view privileged RVFL (FMPRVFL) is proposed based on random vector function link network (RVFL) to accomplish multi-view classification tasks. The basic idea of FMPRVFL lies in the use of additional information from other views on average as the privileged information to supervise the classification in the current view. The objective function of FMPRVFL designed in such a mutually supervised manner can be optimized with analytical solutions, thus accelerating the training of FMPRVFL. It is revealed in the theoretical analysis that in contrast to classical multi-view learning, FMPRVFL can provide extra generalization capability. The results of an experiment on 64 datasets demonstrate that FMPRVFL outperforms other comparative methods both in average testing accuracy and running time.

Key words: multi-view learning (MVL), privileged information, random vector function link network (RVFL)

中图分类号:

TP181

吴天宇, 王士同. 快速多视角特权协同随机向量函数连接网络[J]. 计算机科学与探索, 2022, 16(10): 2320-2329.

WU Tianyu, WANG Shitong. Fast Multi-view Privileged Random Vector Function Link Network[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(10): 2320-2329.

图/表 8

参考文献 28

[1]	PAOY H, PARK G H, SOBAJIC D J. Learning and general-ization characteristics of the random vector functional-link net[J]. Neurocomputing, 1994, 6(2): 163-180. DOI URL
[2]	CHENCL P, WAN J Z. A rapid learning and dynamic step-wise updating algorithm for flat neural networks and the application to time-series prediction[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1999, 29(1): 62-72.
[3]	CHENCL P, LIU Z L. Broad learning system: an effective and efficient incremental learning system without the need for deep architecture[J]. IEEE Transactions on Neural Net-works and Learning Systems, 2018, 29(1): 10-24.
[4]	YANG Y, WANG H. Multi-view clustering: a survey[J]. Big Data Mining and Analytics, 2018, 1(2): 83-107. DOI URL
[5]	XU C, TAO D, XU C. A survey on multi-view learning[J]. arXiv:1304.5634, 2013.
[6]	LI J, ALLINSON N M, TAO D C, et al. Multitraining sup-port vector machine for image retrieval[J]. IEEE Trans-actions on Image Processing, 2006, 15(11): 3597-3601.
[7]	SUN S, SHAWE-TAYLOR J. Sparse semi-supervised learning using conjugate functions[J]. Journal of Machine Learning Research, 2010, 11: 2423-2455.
[8]	SUN S. Multi-view Laplacian support vector machines[C]// LNCS 7121: Proceedings of the 7th International Confer-ence on Advanced Data Mining and Applications, Beijing, Dec 17-19, 2011. Berlin, Heidelberg: Springer, 2011: 209-222.
[9]	LUO Y, TAO D, XU C, et al. Multiview vector-valued mani-fold regularization for multilabel image classification[J]. IEEE Transactions on Neural Networks and Learning Sys-tems, 2013, 24(5): 709-722.
[10]	XU C, TAO D C, XU C. Large-margin multi-view inform-ation bottleneck[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014, 36(8): 1559-1572. DOI URL
[11]	VAPNIK V, VASHIST A. A new learning paradigm: learning using privileged information[J]. Neural Networks, 2009, 22(5/6): 544-557. DOI URL
[12]	VAPNIK V, IZMAILOV R. Learning using privileged infor-mation: similarity control and knowledge transfer[J]. Journal of Machine Learning Research, 2015, 16: 2023-2049.
[13]	姜志彬. 基于多视角学习和迁移学习的分类方法及应用研究[D]. 无锡: 江南大学, 2019.
	JIANG Z B. Classification method and application research based on multi-perspective learning and transfer learning[D]. Wuxi: Jiangnan University, 2019.
[14]	CHENG J, WANG K Q. Multi-view sampling for relevance feedback in image retrieval[C]// Proceedings of the 18th International Conference on Pattern Recognition, Hong Kong, China, Aug 20-24, 2006. Washington: IEEE Computer Society, 2006: 881-884.
[15]	SUN S L, JIN F. Robust co-training[J]. International Journal of Pattern Recognition and Artificial Intelligence, 2011, 25(7): 1113-1126. DOI URL
[16]	ZHAO J, XIE X J, XU X, et al. Multi-view learning over-view: recent progress and new challenges[J]. Information Fusion, 2017, 38: 43-54. DOI URL
[17]	张丹丹, 邓赵红, 王士同. 面向多视角数据的极大熵聚类算法[J]. 计算机科学与探索, 2016, 10(4): 554-564. DOI
	ZHANG D D, DENG Z H, WANG S T. Maximum entropy clustering algorithm for multi-view data[J]. Journal of Frontiers of Computer Science and Technology, 2016, 10(4): 554-564. DOI
[18]	MUSLEA I, MINTON S, KNOBLOCK C A. Active learning with multiple views[J]. Journal of Artificial Intelligence Re-search, 2006, 27: 203-233.
[19]	VALENTE F, WELLEKENS C. Maximum entropy discrim-ination (MED) feature subset selection for speech recog-nition[C]// Proceedings of the 2003 IEEE Workshop on Auto-matic Speech Recognition and Understanding, St Thomas, Nov 30-Dec 4, 2003. Piscataway: IEEE, 2003: 327-332.
[20]	SUN S L, CHAO G Q. Multi-view maximum entropy discrim-ination[C]// Proceedings of the 23rd International Joint Conference on Artificial Intelligence, Beijing, Aug 3-9, 2013. Menlo Park: AAAI, 2013: 1706-1712.
[21]	CHAO G, SUN S. Consensus and complementarity based maximum entropy discrimination for multi-view classific-ation[J]. Information Sciences, 2016, 367: 296-310.
[22]	XU X X, LI W, XU D. Distance metric learning using privileged information for face verification and person re-identification[J]. IEEE Transactions on Neural Networks & Learning Systems, 2015, 26(12): 3150-3162.
[23]	SHI J, XUE Z Y, DAI Y K, et al. Cascaded multi-column RVFL+classifier for single-modal neuroimaging-based diag-nosis of Parkinson’s disease[J]. IEEE Transactions on Bio-medical Engineering, 2018, 66(8): 2362-2371. DOI URL
[24]	BERK R A. Statistical learning from a regression perspe-ctive[M]. Berlin, Heidelberg: Springer, 2008.
[25]	LANGE K, HUNTER D R, YANG I. Optimization transfer using surrogate objective functions[J]. Journal of Computa-tional & Graphical Statistics, 2000, 9(1): 1-20.
[26]	QUEIPO N V, HAFTKA R T, SHYY W, et al. Surrogate-based analysis and optimization[J]. Progress in Aerospace Sciences, 2005, 41(1): 1-28. DOI URL
[27]	XIAN Y Q, LAMPERT C H, SCHIELE B, et al. Zero-shot learning—a comprehensive evaluation of the good, the bad and the ugly[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 41(9): 2251-2265. DOI URL
[28]	CHUA T S, TANG J H, HONG R C, et al. NUS-WIDE: a real-world web image database from National University of Singapore[C]// Proceedings of the 8th ACM International Conference on Image and Video Retrieval, Santorini Island,Jul 8-10, 2009. New York: ACM, 2009: 1-9.

数据集	数量	类别数	特征A	特征B
AwA	6 180	8	2 000(SURF)	252(HOG)
NUS-WIDE	6 265	7	225(CM55)	73(EDH)

数据集	数量	类别数	特征A	特征B
AwA	6 180	8	2 000(SURF)	252(HOG)
NUS-WIDE	6 265	7	225(CM55)	73(EDH)

No.	Dataset A	Dataset B	FMPRVFL		RVFL-A		RVFL-B		SVM-2K		MED-2C		PSVM-2V
No.	Dataset A	Dataset B	Acc/%	STD	Acc/%	STD	Acc/%	STD	Acc/%	STD	Acc/%	STD	Acc/%	STD
1	chimps	panda	90.02	0.008	87.14	0.027	74.61	0.021	86.76	0.028	86.93	0.027	90.92	0.023
2	chimps	leopard	90.72	0.024	86.84	0.018	71.50	0.037	85.73	0.034	82.80	0.040	87.61	0.043
3	chimps	cat	89.24	0.008	86.65	0.034	75.08	0.035	84.31	0.010	82.07	0.049	86.52	0.067
4	chimps	pig	85.13	0.004	82.09	0.050	73.56	0.036	81.88	0.006	82.86	0.033	83.81	0.056
5	chimps	hippo	87.34	0.017	86.84	0.022	71.50	0.037	85.59	0.021	82.82	0.043	84.45	0.063
6	chimps	raccoon	85.61	0.007	83.02	0.020	71.65	0.050	82.23	0.026	79.75	0.062	83.83	0.071
7	chimps	rat	86.15	0.021	82.36	0.041	74.24	0.031	82.24	0.108	75.12	0.035	81.16	0.082
8	chimps	seal	88.04	0.027	81.46	0.035	76.79	0.021	88.43	0.104	83.07	0.026	87.76	0.016
9	panda	leopard	91.49	0.009	88.84	0.014	69.05	0.042	88.95	0.019	84.31	0.026	90.46	0.012
10	panda	cat	92.31	0.004	90.04	0.026	78.33	0.021	89.63	0.027	88.03	0.020	89.53	0.067
11	panda	pig	87.09	0.020	83.62	0.025	76.02	0.027	84.08	0.020	78.83	0.042	83.46	0.035
12	panda	hippo	90.42	0.018	87.69	0.018	79.46	0.025	88.48	0.017	87.47	0.009	91.13	0.050
13	panda	raccoon	90.44	0.027	86.90	0.030	69.97	0.039	89.56	0.027	88.94	0.024	90.85	0.037
14	panda	rat	88.12	0.028	85.92	0.043	81.33	0.025	82.76	0.024	82.25	0.025	86.57	0.064
15	panda	seal	90.13	0.025	89.10	0.017	83.17	0.037	88.06	0.033	86.89	0.033	89.63	0.026
16	leopard	cat	90.80	0.017	86.94	0.021	84.37	0.028	86.84	0.026	86.15	0.033	90.69	0.045
17	leopard	pig	83.05	0.022	82.62	0.028	73.41	0.064	82.19	0.037	78.53	0.027	84.11	0.057
18	leopard	hippo	87.60	0.017	83.10	0.028	75.26	0.038	86.06	0.036	82.54	0.037	88.49	0.092
19	leopard	raccoon	82.79	0.014	78.25	0.032	61.44	0.030	77.19	0.042	74.54	0.041	81.21	0.074
20	leopard	rat	86.15	0.029	84.53	0.033	73.12	0.070	82.93	0.052	80.17	0.040	85.21	0.082
21	leopard	seal	90.63	0.006	86.47	0.027	78.75	0.028	88.01	0.033	87.83	0.045	88.00	0.091
22	cat	pig	81.75	0.035	79.27	0.044	75.20	0.047	75.68	0.016	73.92	0.061	76.80	0.055
23	cat	hippo	88.26	0.015	87.08	0.029	75.63	0.037	85.37	0.024	85.07	0.037	86.29	0.074
24	cat	raccoon	87.14	0.014	84.76	0.021	65.44	0.023	88.58	0.029	85.06	0.022	89.85	0.043
25	cat	rat	75.26	0.032	74.17	0.040	69.12	0.023	68.42	0.027	62.40	0.025	68.46	0.043
26	cat	seal	84.85	0.032	76.55	0.044	71.32	0.048	82.10	0.023	82.60	0.038	83.68	0.060
27	pig	hippo	78.94	0.050	77.21	0.055	71.49	0.045	73.79	0.023	71.42	0.037	74.49	0.066
28	pig	raccoon	81.97	0.018	78.30	0.034	69.47	0.059	81.09	0.038	76.75	0.016	79.46	0.065
29	pig	rat	73.18	0.029	69.21	0.032	60.43	0.069	71.43	0.143	70.52	0.025	74.31	0.080
30	pig	seal	80.43	0.045	78.27	0.045	75.71	0.023	77.02	0.020	71.86	0.037	77.08	0.092
31	hippo	raccoon	85.40	0.013	83.49	0.030	75.52	0.038	82.60	0.032	80.54	0.045	83.53	0.036
32	hippo	rat	84.81	0.011	82.92	0.019	76.67	0.021	76.07	0.018	72.94	0.035	75.23	0.056
33	hippo	seal	71.93	0.034	70.45	0.034	68.26	0.015	64.46	0.014	67.08	0.030	69.48	0.053
34	raccoon	rat	79.88	0.027	78.45	0.039	69.55	0.017	75.23	0.038	74.14	0.026	78.76	0.024
35	raccoon	seal	89.49	0.012	85.11	0.019	77.83	0.031	87.30	0.043	84.08	0.040	90.72	0.019
36	rat	seal	80.59	0.036	77.32	0.042	72.02	0.042	73.81	0.018	71.96	0.020	75.69	0.017
	Average		85.48	0.021	82.58	0.031	73.51	0.036	82.08	0.034	79.78	0.034	83.59	0.054

No.	Dataset A	Dataset B	FMPRVFL		RVFL-A		RVFL-B		SVM-2K		MED-2C		PSVM-2V
No.	Dataset A	Dataset B	Acc/%	STD	Acc/%	STD	Acc/%	STD	Acc/%	STD	Acc/%	STD	Acc/%	STD
1	chimps	panda	90.02	0.008	87.14	0.027	74.61	0.021	86.76	0.028	86.93	0.027	90.92	0.023
2	chimps	leopard	90.72	0.024	86.84	0.018	71.50	0.037	85.73	0.034	82.80	0.040	87.61	0.043
3	chimps	cat	89.24	0.008	86.65	0.034	75.08	0.035	84.31	0.010	82.07	0.049	86.52	0.067
4	chimps	pig	85.13	0.004	82.09	0.050	73.56	0.036	81.88	0.006	82.86	0.033	83.81	0.056
5	chimps	hippo	87.34	0.017	86.84	0.022	71.50	0.037	85.59	0.021	82.82	0.043	84.45	0.063
6	chimps	raccoon	85.61	0.007	83.02	0.020	71.65	0.050	82.23	0.026	79.75	0.062	83.83	0.071
7	chimps	rat	86.15	0.021	82.36	0.041	74.24	0.031	82.24	0.108	75.12	0.035	81.16	0.082
8	chimps	seal	88.04	0.027	81.46	0.035	76.79	0.021	88.43	0.104	83.07	0.026	87.76	0.016
9	panda	leopard	91.49	0.009	88.84	0.014	69.05	0.042	88.95	0.019	84.31	0.026	90.46	0.012
10	panda	cat	92.31	0.004	90.04	0.026	78.33	0.021	89.63	0.027	88.03	0.020	89.53	0.067
11	panda	pig	87.09	0.020	83.62	0.025	76.02	0.027	84.08	0.020	78.83	0.042	83.46	0.035
12	panda	hippo	90.42	0.018	87.69	0.018	79.46	0.025	88.48	0.017	87.47	0.009	91.13	0.050
13	panda	raccoon	90.44	0.027	86.90	0.030	69.97	0.039	89.56	0.027	88.94	0.024	90.85	0.037
14	panda	rat	88.12	0.028	85.92	0.043	81.33	0.025	82.76	0.024	82.25	0.025	86.57	0.064
15	panda	seal	90.13	0.025	89.10	0.017	83.17	0.037	88.06	0.033	86.89	0.033	89.63	0.026
16	leopard	cat	90.80	0.017	86.94	0.021	84.37	0.028	86.84	0.026	86.15	0.033	90.69	0.045
17	leopard	pig	83.05	0.022	82.62	0.028	73.41	0.064	82.19	0.037	78.53	0.027	84.11	0.057
18	leopard	hippo	87.60	0.017	83.10	0.028	75.26	0.038	86.06	0.036	82.54	0.037	88.49	0.092
19	leopard	raccoon	82.79	0.014	78.25	0.032	61.44	0.030	77.19	0.042	74.54	0.041	81.21	0.074
20	leopard	rat	86.15	0.029	84.53	0.033	73.12	0.070	82.93	0.052	80.17	0.040	85.21	0.082
21	leopard	seal	90.63	0.006	86.47	0.027	78.75	0.028	88.01	0.033	87.83	0.045	88.00	0.091
22	cat	pig	81.75	0.035	79.27	0.044	75.20	0.047	75.68	0.016	73.92	0.061	76.80	0.055
23	cat	hippo	88.26	0.015	87.08	0.029	75.63	0.037	85.37	0.024	85.07	0.037	86.29	0.074
24	cat	raccoon	87.14	0.014	84.76	0.021	65.44	0.023	88.58	0.029	85.06	0.022	89.85	0.043
25	cat	rat	75.26	0.032	74.17	0.040	69.12	0.023	68.42	0.027	62.40	0.025	68.46	0.043
26	cat	seal	84.85	0.032	76.55	0.044	71.32	0.048	82.10	0.023	82.60	0.038	83.68	0.060
27	pig	hippo	78.94	0.050	77.21	0.055	71.49	0.045	73.79	0.023	71.42	0.037	74.49	0.066
28	pig	raccoon	81.97	0.018	78.30	0.034	69.47	0.059	81.09	0.038	76.75	0.016	79.46	0.065
29	pig	rat	73.18	0.029	69.21	0.032	60.43	0.069	71.43	0.143	70.52	0.025	74.31	0.080
30	pig	seal	80.43	0.045	78.27	0.045	75.71	0.023	77.02	0.020	71.86	0.037	77.08	0.092
31	hippo	raccoon	85.40	0.013	83.49	0.030	75.52	0.038	82.60	0.032	80.54	0.045	83.53	0.036
32	hippo	rat	84.81	0.011	82.92	0.019	76.67	0.021	76.07	0.018	72.94	0.035	75.23	0.056
33	hippo	seal	71.93	0.034	70.45	0.034	68.26	0.015	64.46	0.014	67.08	0.030	69.48	0.053
34	raccoon	rat	79.88	0.027	78.45	0.039	69.55	0.017	75.23	0.038	74.14	0.026	78.76	0.024
35	raccoon	seal	89.49	0.012	85.11	0.019	77.83	0.031	87.30	0.043	84.08	0.040	90.72	0.019
36	rat	seal	80.59	0.036	77.32	0.042	72.02	0.042	73.81	0.018	71.96	0.020	75.69	0.017
	Average		85.48	0.021	82.58	0.031	73.51	0.036	82.08	0.034	79.78	0.034	83.59	0.054

No.	Dataset A	Dataset B	FMPRVFL		RVFL-A		RVFL-B		SVM-2K		MED-2C		PSVM-2V
No.	Dataset A	Dataset B	Acc/%	STD	Acc/%	STD	Acc/%	STD	Acc/%	STD	Acc/%	STD	Acc/%	STD
1	birds	computer	83.70	0.018	81.50	0.017	81.83	0.024	81.30	0.021	75.10	0.033	79.41	0.020
2	birds	flowers	82.34	0.012	74.74	0.014	79.81	0.012	81.48	0.017	81.12	0.026	82.27	0.016
3	birds	lake	75.63	0.011	74.67	0.014	70.60	0.023	68.00	0.014	68.08	0.037	72.43	0.012
4	birds	plants	78.04	0.017	70.56	0.022	77.53	0.014	78.95	0.010	74.56	0.036	79.76	0.016
5	birds	sign	77.19	0.029	75.83	0.027	69.74	0.035	67.13	0.009	67.59	0.042	71.66	0.026
6	birds	swimmers	90.05	0.013	86.39	0.008	85.66	0.011	87.58	0.030	86.83	0.042	88.56	0.028
7	birds	vehicle	83.96	0.018	82.56	0.006	80.56	0.015	81.83	0.021	89.74	0.052	83.17	0.026
8	computer	flowers	91.19	0.012	90.71	0.009	91.04	0.005	90.09	0.009	89.74	0.074	90.57	0.008
9	computer	lake	78.41	0.022	76.99	0.016	70.20	0.025	75.46	0.020	71.74	0.035	75.95	0.039
10	computer	plants	79.64	0.036	75.95	0.026	77.36	0.019	76.98	0.032	78.29	0.011	78.62	0.016
11	computer	sign	76.48	0.027	73.72	0.020	72.13	0.024	75.04	0.037	73.86	0.010	75.38	0.026
12	computer	swimmers	78.68	0.033	75.53	0.025	72.62	0.074	75.04	0.037	76.77	0.008	79.60	0.034
13	computer	vehicle	74.14	0.033	73.53	0.025	72.62	0.074	66.46	0.031	60.39	0.010	69.06	0.037
14	flowers	lake	85.75	0.004	84.47	0.006	83.29	0.008	83.87	0.008	83.72	0.008	84.12	0.013
15	flowers	plants	84.07	0.019	82.07	0.019	81.10	0.019	83.23	0.022	83.08	0.034	83.18	0.010
16	flowers	sign	85.07	0.008	81.83	0.020	76.95	0.013	77.39	0.014	77.67	0.011	83.18	0.010
17	flowers	swimmers	94.93	0.004	93.58	0.008	93.52	0.010	93.51	0.011	93.91	0.017	94.43	0.008
18	flowers	vehicle	92.29	0.010	89.74	0.010	89.74	0.010	91.50	0.006	90.19	0.019	91.32	0.011
19	lake	plants	80.62	0.019	78.21	0.015	77.89	0.018	79.33	0.026	79.26	0.021	79.74	0.029
20	lake	sign	77.08	0.011	75.78	0.008	67.95	0.027	71.35	0.017	68.84	0.026	72.33	0.017
21	lake	swimmers	84.00	0.035	82.57	0.027	80.36	0.031	80.52	0.019	81.91	0.018	84.16	0.011
22	lake	vehicle	79.96	0.035	77.43	0.035	78.46	0.031	79.67	0.021	76.46	0.009	81.51	0.034
23	plants	sign	78.89	0.022	77.60	0.018	73.93	0.029	78.23	0.026	78.88	0.036	76.71	0.018
24	plants	swimmers	83.81	0.032	80.03	0.038	79.14	0.044	81.68	0.037	82.40	0.009	84.07	0.009
25	plants	vehicle	81.61	0.026	80.75	0.016	80.49	0.026	81.73	0.041	79.54	0.014	80.56	0.036
26	sign	swimmers	87.39	0.027	85.11	0.014	84.34	0.012	86.47	0.028	84.74	0.036	85.70	0.021
27	sign	swimmers	81.00	0.014	80.21	0.019	77.31	0.029	77.47	0.007	76.16	0.027	78.08	0.014
28	swimmers	vehicle	80.30	0.032	75.50	0.018	74.05	0.013	76.99	0.045	74.05	0.016	78.08	0.014
	Average		82.37	0.021	79.91	0.018	78.58	0.024	79.58	0.022	78.74	0.026	80.84	0.020

快速多视角特权协同随机向量函数连接网络

Fast Multi-view Privileged Random Vector Function Link Network

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 28

相关文章 15

编辑推荐

Metrics

[1]	刘学文, 王继奎, 杨正国, 李冰, 聂飞平. 密度峰值隶属度优化的半监督Self-Training算法[J]. 计算机科学与探索, 2022, 16(9): 2078-2088.
[2]	谢娟英, 张凯云. XR-MSF-Unet：新冠肺炎肺部CT图像自动分割模型[J]. 计算机科学与探索, 2022, 16(8): 1850-1864.
[3]	陈洋, 王士同. 多样性正则化极限学习机的集成方法[J]. 计算机科学与探索, 2022, 16(8): 1819-1928.
[4]	杨政, 邓赵红, 罗晓清, 顾鑫, 王士同. 利用ELM-AE和迁移表征学习构建的目标跟踪系统[J]. 计算机科学与探索, 2022, 16(7): 1633-1648.
[5]	董文波, 孙仕亮, 殷敏智. 医学知识推理研究现状与发展[J]. 计算机科学与探索, 2022, 16(6): 1193-1213.
[6]	申瑞彩, 翟俊海, 侯璎真. 选择性集成学习多判别器生成对抗网络[J]. 计算机科学与探索, 2022, 16(6): 1429-1438.
[7]	谢欣彤, 胡悦阳, 刘譞哲, 赵耀帅, 姜海鸥. 传播用户代表性特征学习的谣言检测方法[J]. 计算机科学与探索, 2022, 16(6): 1334-1342.
[8]	张壮, 王士同. 不平衡数据的Takagi-Sugeno-Kang模糊分类集成模型[J]. 计算机科学与探索, 2022, 16(6): 1374-1382.
[9]	林佳伟, 王士同. 用于无监督域适应的深度对抗重构分类网络[J]. 计算机科学与探索, 2022, 16(5): 1107-1116.
[10]	孙武, 邓赵红, 娄琼丹, 顾鑫, 王士同. 基于模糊规则学习的无监督异构领域自适应[J]. 计算机科学与探索, 2022, 16(2): 403-412.
[11]	章枫叶欣, 王骏, 贾修一, 潘祥, 邓赵红, 施俊, 王士同. 面向多分类自闭症辅助诊断的标记分布学习[J]. 计算机科学与探索, 2022, 16(1): 194-204.
[12]	付彬, 王志海, 王中锋. 最大化边际的分类器选取算法 [J]. 计算机科学与探索, 2011, 5(1): 59-67.
[13]	潘世瑞1 , 张阳1,2+ , 李雪 3 , 王勇 4 . 针对不确定正例和未标记学习的最近邻算法*[J]. 计算机科学与探索, 2010, 4(9): 769-779.
[14]	袁磊1 , 张阳2+ , 李梅1 , 李雪3 , 王勇4 . 在数据流管理系统中实现快速决策树算法*[J]. 计算机科学与探索, 2010, 4(8): 673-682.
[15]	石国强, 牛常勇, 范明+ . 使用PCA建立基于规则的组合分类器*[J]. 计算机科学与探索, 2010, 4(5): 455-463.