×

Some theoretical results of learning theory based on random sets in set-valued probability space. (English) Zbl 1197.93166

Summary: The purpose of this paper is to introduce some basic knowledge of Statistical Learning Theory (SLT) based on random set samples in set-valued probability space for the first time and generalize the key theorem and bounds on the rate of uniform convergence of learning theory in Vapnik, to the key theorem and bounds on the rate of uniform convergence for random sets in set-valued probability space. SLT based on random samples formed in probability space is considered, at present, as one of the fundamental theories about small samples statistical learning. It has become a novel and important field of machine learning, along with other concepts and architectures such as neural networks. However, the theory hardly handles statistical learning problems for samples that involve random set samples.
Being motivated by some applications, in this paper a SLT is developed based on random set samples. First, a certain law of large numbers for random sets is proved. Second, the definitions of the distribution function and the expectation of random sets are introduced, and the concepts of the expected risk functional and the empirical risk functional are discussed. A notion of the strict consistency of the principle of empirical risk minimization is presented.
The paper formulates and proves the key theorem and presents the bounds on the rate of uniform convergence of learning theory based on random sets in set-valued probability space, which become cornerstones of the theoretical fundamentals of the SLT for random set samples.
The paper provides a studied analysis of some theoretical results of learning theory.

MSC:

93E35 Stochastic learning and adaptive control
93E03 Stochastic systems in control theory (general)
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] DOI: 10.1090/S0002-9947-1972-0293054-4 · doi:10.1090/S0002-9947-1972-0293054-4
[2] DOI: 10.1016/0022-247X(65)90049-1 · Zbl 0163.06301 · doi:10.1016/0022-247X(65)90049-1
[3] DOI: 10.1109/72.788648 · doi:10.1109/72.788648
[4] DOI: 10.1016/0047-259X(78)90022-2 · Zbl 0384.28006 · doi:10.1016/0047-259X(78)90022-2
[5] DOI: 10.1016/j.fss.2003.06.011 · Zbl 1104.60307 · doi:10.1016/j.fss.2003.06.011
[6] DOI: 10.1016/0022-247X(90)90200-Y · Zbl 0721.60056 · doi:10.1016/0022-247X(90)90200-Y
[7] DOI: 10.1016/S0362-546X(03)00190-1 · Zbl 1022.49018 · doi:10.1016/S0362-546X(03)00190-1
[8] DOI: 10.1287/moor.11.3.385 · Zbl 0611.60004 · doi:10.1287/moor.11.3.385
[9] DOI: 10.1016/S1631-073X(02)02382-8 · Zbl 1022.28006 · doi:10.1016/S1631-073X(02)02382-8
[10] DOI: 10.1016/S1631-073X(02)00027-4 · Zbl 1041.28010 · doi:10.1016/S1631-073X(02)00027-4
[11] DOI: 10.1016/S0305-0483(01)00026-3 · doi:10.1016/S0305-0483(01)00026-3
[12] Tsong, J.J. (2006), ”Hybrid approach of selecting hyper-parameters of support vector machine for regression”,IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, Vol. 36 No. 3, pp. 699-709.
[13] DOI: 10.2307/2525560 · Zbl 0126.36401 · doi:10.2307/2525560
[14] Yong, G. and Zhang, W.X. (1994), ”Theory of selection operators on hyperspaces and set-valued stochastic processes”,Science in China, Vol. 24 No. 2, pp. 113-21.
[15] DOI: 10.1016/0165-0114(90)90091-J · Zbl 0698.28010 · doi:10.1016/0165-0114(90)90091-J
[16] DOI: 10.1016/j.comnet.2004.09.004 · Zbl 1067.68511 · doi:10.1016/j.comnet.2004.09.004
[17] DOI: 10.1016/S0167-9473(01)00069-X · Zbl 1072.62642 · doi:10.1016/S0167-9473(01)00069-X
[18] Guergachi, A.A. (2001), ”Using statistical learning theory for modeling the uncertainty in business and engineering systems: a qualitative introduction”,Systems, Man, and Cybernetics, 2001 IEEE International Conference, pp. 423-8. · doi:10.1109/ICSMC.2001.969849
[19] DOI: 10.1109/72.788640 · doi:10.1109/72.788640
[20] DOI: 10.1109/TPAMI.2004.1265862 · Zbl 05111088 · doi:10.1109/TPAMI.2004.1265862
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.