Corpus ID: 11928273

Maximum a Posteriori Estimation by Search in Probabilistic Programs

@article{Tolpin2015MaximumAP,
  title={Maximum a Posteriori Estimation by Search in Probabilistic Programs},
  author={David Tolpin and F. Wood},
  journal={ArXiv},
  year={2015},
  volume={abs/1504.06848}
}
  • David Tolpin, F. Wood
  • Published 2015
  • Computer Science, Mathematics
  • ArXiv
  • We introduce an approximate search algorithm for fast maximum a posteriori probability estimation in probabilistic programs, which we call Bayesian ascent Monte Carlo (BaMC). Probabilistic programs represent probabilistic models with varying number of mutually dependent finite, countable, and continuous random variables. BaMC is an anytime MAP search algorithm applicable to any combination of random variables and dependencies. We compare BaMC to other MAP estimation algorithms and show that… CONTINUE READING

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 57 REFERENCES
    Probabilistic Deterministic Infinite Automata
    19
    Solving MAP Exactly using Systematic Search
    66
    Simulated annealing for maximum a Posteriori parameter estimation of hidden Markov models
    38
    Marginal maximum a posteriori estimation using Markov chain Monte Carlo
    92
    Asynchronous Anytime Sequential Monte Carlo
    41
    Lightweight Implementations of Probabilistic Programming Languages Via Transformational Compilation
    144
    Bayesian Mixture Modelling and Inference based Thompson Sampling in Monte-Carlo Tree Search
    19
    Bandit Based Monte-Carlo Planning
    2204
    Convergence of an annealing algorithm
    491