Wild Variational Approximations

نویسندگان

  • Yingzhen Li
  • Qiang Liu
چکیده

We formalise the research problem of approximate inference in the wild: developing new variants of variational methods that work for arbitrary variational approximation families for which inference (e.g., sampling or calculating expectation) is tractable, but probability density function may be intractable. We provide several examples for this type of approximations, discuss energy/gradient approximation for existing techniques, and further comment on developing other variational objectives and amortising stochastic dynamics. Connections and comparisons to existing approaches are also briefly discussed.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Approximate Solution of Fuzzy Fractional Differential Equations

‎In this paper we propose a method for computing approximations of solution of fuzzy fractional differential equations using fuzzy variational iteration method. Defining a fuzzy fractional derivative, we verify the utility of the method through two illustrative ‎examples.‎

متن کامل

Explaining Variational Approximations

Variational approximations facilitate approximate inference for the parameters in complex statistical models and provide fast, deterministic alternatives to Monte Carlo methods. However, much of the contemporary literature on variational approximations is in Computer Science rather than Statistics, and uses terminology, notation, and examples from the former field. In this article we explain va...

متن کامل

A New Modification of the Reconstruction of Variational Iteration Method for Solving Multi-order Fractional Differential Equations

Fractional calculus has been used to model the physical and engineering processes that have found to be best described by fractional differential equations. For that reason, we need a reliable and efficient technique for the solution of fractional differential equations. The aim of this paper is to present an analytical approximation solution for linear and nonlinear multi-order fractional diff...

متن کامل

Improved Variational Approximation for Bayesian PCA

As with most non-trivial models, an exact Bayesian treatment of the probabilistic PCA model (under a meaningful prior) is analytically intractable. Various approximations have therefore been proposed in the literature; these include approximations based on type-II maximum likelihood as well as variational approximations. In this document, we describe an improved variational approximation for Ba...

متن کامل

A new numerical scheme for solving systems of integro-differential equations

This paper has been devoted to apply the Reconstruction of Variational Iteration Method (RVIM) to handle the systems of integro-differential equations. RVIM has been induced with Laplace transform from the variational iteration method (VIM) which was developed from the Inokuti method. Actually, RVIM overcome to shortcoming of VIM method to determine the Lagrange multiplier. So that, RVIM method...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2016