Some variance reduction methods for numerical stochastic homogenization

نویسندگان
چکیده

برای دانلود باید عضویت طلایی داشته باشید

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

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Stochastic Variance Reduction Methods for Policy Evaluation

Policy evaluation is a crucial step in many reinforcement-learning procedures, which estimates a value function that predicts states’ longterm value under a given policy. In this paper, we focus on policy evaluation with linear function approximation over a fixed dataset. We first transform the empirical policy evaluation problem into a (quadratic) convex-concave saddle point problem, and then ...

متن کامل

Variance Reduction in Stochastic Homogenization Using Antithetic Variables

Some theoretical issues related to the problem of variance reduction in numerical approaches for stochastic homogenization are examined. On some simple, yet representative cases, it is demonstrated theoretically that a technique based on antithetic variables can indeed reduce the variance of the output of the computation, and decrease the overall computational cost of such a multiscale problem....

متن کامل

Numerical homogenization methods

Numerical homogenization methods Synonyms multiscale methods for homogenization problems, upscaling methods, representative volume element methods Definition Numerical homogenization methods are techniques for finding numerical solutions of partial differential equations (PDEs) with rapidly oscillating coefficients (multiple scales). In mathematical analysis, homogenization can be defined as a ...

متن کامل

Stochastic Variance Reduction Methods for Saddle-Point Problems

We consider convex-concave saddle-point problems where the objective functionsmay be split in many components, and extend recent stochastic variance reductionmethods (such as SVRG or SAGA) to provide the first large-scale linearly conver-gent algorithms for this class of problems which are common in machine learning.While the algorithmic extension is straightforward, it come...

متن کامل

Some Numerical Approaches for Weakly Random Homogenization

We overview a series of recent works addressing homogenization problems for some materials seen as small random perturbations of periodic materials (in a sense made precise in the body of the text). These recent works are joint works with several collaborators: Blanc (Paris 6), Lions (Collège de France), Legoll, Anantharaman, Costaouec (Ecole Nationale des Ponts et Chaussées and INRIA). The the...

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

ژورنال

عنوان ژورنال: Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences

سال: 2016

ISSN: 1364-503X,1471-2962

DOI: 10.1098/rsta.2015.0168