相关向量机

相关向量机（Relevance vector machine，RVM）是使用贝叶斯推理得到回归和分类的解的机器学习技术。RVM的函数形式与支持向量机相同，但是可以提供概率分类。 其与带协方差函数的高斯过程等效。： 其中φ是（通常是高斯核函数），x1,…,xN是训练集的输入向量。 Compared to the SVM the Bayesian formulation allows avoiding the set of free parameters that the SVM has and that usually require cross-validation based post optimizations. However RVMs use an (EM)-like learning method and are therefore at risk of local minima, unlike the standard -based algorithms employed by s which are guaranteed to find a global optimum.