Matt Hoffman 0001 Arnaud Doucet Nando de Freitas Ajay Jasra Bayesian Policy Learning with Trans-Dimensional MCMC. 2007 NIPS https://proceedings.neurips.cc/paper/2007/hash/3a15c7d0bbe60300a39f76f8a5ba6896-Abstract.html http://papers.nips.cc/paper/3343-bayesian-policy-learning-with-trans-dimensional-mcmc conf/nips/2007 db/conf/nips/nips2007.html#HoffmanDFJ07 665-672