TY - JOUR
T1 - Parameter-insensitive kernel in extreme learning for non-linear support vector regression
AU - Frénay, Benoît
AU - Verleysen, Michel
N1 - M1 - 16
PY - 2011/9
Y1 - 2011/9
N2 - Support vector regression (SVR) is a state-of-the-art method for regression which uses the εsensitive loss and produces sparse models. However, non-linear SVRs are difficult to tune because of the additional kernel parameter. In this paper, a new parameter-insensitive kernel inspired from extreme learning is used for non-linear SVR. Hence, the practitioner has only two meta-parameters to optimise. The proposed approach reduces significantly the computational complexity yet experiments show that it yields performances that are very close from the state-of-the-art. Unlike previous works which rely on Monte-Carlo approximation to estimate the kernel, this work also shows that the proposed kernel has an analytic form which is computationally easier to evaluate. © 2011 Elsevier B.V.
AB - Support vector regression (SVR) is a state-of-the-art method for regression which uses the εsensitive loss and produces sparse models. However, non-linear SVRs are difficult to tune because of the additional kernel parameter. In this paper, a new parameter-insensitive kernel inspired from extreme learning is used for non-linear SVR. Hence, the practitioner has only two meta-parameters to optimise. The proposed approach reduces significantly the computational complexity yet experiments show that it yields performances that are very close from the state-of-the-art. Unlike previous works which rely on Monte-Carlo approximation to estimate the kernel, this work also shows that the proposed kernel has an analytic form which is computationally easier to evaluate. © 2011 Elsevier B.V.
KW - ELM kernel
KW - Extreme learning machine
KW - Infinite number of neurons
KW - Support vector regression
UR - http://www.scopus.com/inward/record.url?scp=80051670315&partnerID=8YFLogxK
U2 - 10.1016/j.neucom.2010.11.037
DO - 10.1016/j.neucom.2010.11.037
M3 - Article
SN - 0925-2312
VL - 74
SP - 2526
EP - 2531
JO - Neurocomputing
JF - Neurocomputing
IS - 16
ER -