Tipping, M. E. (2001). Sparse Bayesian learning and the relevance vector machine. Journal of Machine Learning Research 1, 211–244. [Abstract] [Available from JMLR]
-
Page 219, equation (27) should have a \hat{t} in place of the t, where
\hat{t} = Fw + B-1(t-y)
-
Page 226, Section 4.4.1, Regression results table: for both 'sinc' examples, quoted values for "errors" for both RVM and SVM are missing a zero immediately after the decimal point. i.e. that section of the table should read:
0.0378
0.0326
0.0215
0.0187
-
Page 239, equation (60). CNGP should be replaced by its inverse.