A conditional game for comparing approximations
نویسنده
چکیده
We present a “conditional game” to be played between two approximate inference algorithms. We prove that exact inference is an optimal strategy and demonstrate how the game can be used to estimate the relative accuracy of two different approximations in the absence of exact marginals.
منابع مشابه
Discussion of "A conditional game for comparing approximations"
This brief paper discusses the paper by Eaton mentioned in the title [1].
متن کاملMoment Generating Functions, Local Approximations and One-Step Conditioning
This short note considers how values of ratios of moment generating functions cleanly arise when comparing local approximations to probability functions for convolutions to those resulting through one-step conditioning. In particular, for an independent and identically distributed sequence {Xi} and a sequence {p̃i} which approximates the probability function for the sum ∑ 1<i<n Xi, we compare p̃i...
متن کاملAssessing the Performance of the Ensemble Kalman Filter for Land Surface Data Assimilation
The ensemble Kalman filter provides an easy-to-use, flexible, and efficient option for data assimilation problems. One of its attractive features in land surface applications is its ability to provide distributional information about variables, such as soil moisture, that can be highly skewed or even bimodal. The ensemble Kalman filter relies on normality approximations that improve its efficie...
متن کاملComparing the asymptotic and empirical (un)conditional distributions of OLS and IV in a linear static simultaneous equation
In designing Monte Carlo simulation studies for analyzing nite sample properties of econometric inference methods, one can use either IID drawings in each replication for any series of exogenous explanatory variables or condition on just one realization of these. The results will usually di¤er, as do their interpretations. Conditional and unconditional limiting distributions are often equivale...
متن کاملCOMPARING NUMERICAL METHODS FOR THE SOLUTION OF THE DAMPED FORCED OSCILLATOR PROBLEM
In this paper, we present a comparative study between the Adomian decomposition method and two classical well-known Runge-Kutta and central difference methods for the solution of damped forced oscillator problem. We show that the Adomian decomposition method for this problem gives more accurate approximations relative to other numerical methods and is easier to apply.
متن کامل