Bayesian Learning

Bayesian Learning (BL) is a process whereby a machine is trained on past data so it can come to decisions about the future. These decisions - that something will happen, or that something is true or false - are probabilistic and are never a surety, but more a recommendation. Speciality Bayesian learner can make predicitions incrementally, that is, as more knowledge is fed into the system, the System's prediction changes instantenously. Bayesian Belief Networks (BBN) are not 'black boxes' like Neural Networks (NN) , where no one can know what's going on in the system. NN generally give yes or no answers, while a probalistic estimate is oftern preferable. With BBN, the 'answer' is updated as soon as data is updated, which is not the case with NN.
Proposer :- English theologian Thomas Bayes