P. Sollich (1999)
Probabilistic interpretation and Bayesian methods for Support Vector Machines
In: Proceedings of ICANN'99, pp. 91-96, IEE Publications.
SVMs can be interpreted as MAP-solutions to inference problems with Gaussian Process priors. I show how this helps with the intuitive interpretation of SVM kernels; it also allows Bayesian methods to be used for SVMs (tuning hyper-parameters by maximizing evidence, defining error bars etc).