Discriminative MCMC

Reference:

Kai Puolam?i, Jarkko Salojarvi, Eerika Savia, and Samuel Kaski. Discriminative MCMC. Publications in Computer and Information Science E1, Helsinki University of Technology, Espoo, Finland, 2006.

Abstract:

We discuss Bayesian modeling in the case where the model is incorrect. Standard posterior distribution is optimal for inference if the true model is within the model family. In the case of an incorrect model, we show that for inference on conditioned distribution, a different posterior-type distribution is optimal. We provide here an axiomatic justification of previously suggested supervised posterior distribution, introduce Markov Chain Monte Carlo -type methods for computing with the posterior, and demonstrate empirically that it works as expected.

Suggested BibTeX entry:

@techreport{Puolamaki06,
    address = {Espoo, Finland},
    author = {Kai Puolam?i and Jarkko Salojarvi and Eerika Savia and Samuel Kaski},
    institution = {Helsinki University of Technology},
    number = {E1},
    title = {Discriminative {MCMC}},
    type = {Publications in Computer and Information Science},
    year = {2006},
}