A stable manifold MCMC method for high dimensions
نویسندگان
چکیده
منابع مشابه
A Stable Manifold MCMC Method for High Dimensions
We combine two important recent advancements of MCMC algorithms: first, methods utilizing the intrinsic manifold structure of the parameter space; then, algorithms effective for targets in infinite-dimensions with the critical property that their mixing time is robust to mesh refinement.
متن کاملSingular manifold method for an equation in 2 + 1 dimensions
The Singular Manifold Method is presented as an excellent tool to study a 2 + 1 dimensional equation in despite of the fact that the same method presents several problems when applied to 1 + 1 reductions of the same equation. Nevertheless these problems are solved when the number of dimensions of the equation is increased.
متن کاملThe Cangemi-Jackiw manifold in high dimensions and symplectic structure
The notion of Poincaré gauge manifold (G), proposed in the context of an (1 + 1) gravitational theory by Cangemi and Jackiw ( D. Cangemi and R. Jackiw, Ann. Phys. (N.Y.) 225 (1993) 229), is explored from a geometrical point of view. First G is defined for arbitrary dimensions and in the sequence a symplectic structure is attached to T ∗G. Treating then the case of 5-dimensions, a (4, 1) de-Sitt...
متن کاملCPSG-MCMC: Clustering-Based Preprocessing method for Stochastic Gradient MCMC
In recent years, stochastic gradient Markov Chain Monte Carlo (SG-MCMC) methods have been raised to process large-scale dataset by iterative learning from small minibatches. However, the high variance caused by naive subsampling usually slows down the convergence to the desired posterior distribution. In this paper, we propose an effective subsampling strategy to reduce the variance based on a ...
متن کاملEvaluation of the Centre Manifold Method for Limit Cycle Calculations of a Nonlinear Structural Wing
In this study the centre manifold is applied for reduction and limit cycle calculation of a highly nonlinear structural aeroelastic wing. The limit cycle is arisen from structural nonlinearity due to the large deflection of the wing. Results obtained by different orders of centre manifolds are compared with those obtained by time marching method (fourth-order Runge-Kutta method). These comparis...
متن کاملذخیره 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: Statistics & Probability Letters
سال: 2014
ISSN: 0167-7152
DOI: 10.1016/j.spl.2014.03.016