Weighted ensemble: Recent mathematical developments

06/29/2022
by   D. Aristoff, et al.
0

The weighted ensemble (WE) method, an enhanced sampling approach based on periodically replicating and pruning trajectories in a set of parallel simulations, has grown increasingly popular for computational biochemistry problems, due in part to improved hardware and the availability of modern software. Algorithmic and analytical improvements have also played an important role, and progress has accelerated in recent years. Here, we discuss and elaborate on the WE method from a mathematical perspective, highlighting recent results which have begun to yield greater computational efficiency. Notable among these innovations are variance reduction approaches that optimize trajectory management for systems of arbitrary dimensionality.

READ FULL TEXT
research
02/21/2023

Classy Ensemble: A Novel Ensemble Algorithm for Classification

We present Classy Ensemble, a novel ensemble-generation algorithm for cl...
research
06/10/2018

Dissipativity Theory for Accelerating Stochastic Variance Reduction: A Unified Analysis of SVRG and Katyusha Using Semidefinite Programs

Techniques for reducing the variance of gradient estimates used in stoch...
research
03/15/2022

GBEM: Galerkin Boundary Element Method for 3-D Capacitance Extraction

For modern IC design, electromagnetic coupling among interconnect wires ...
research
02/24/2022

A Note on Machine Learning Approach for Computational Imaging

Computational imaging has been playing a vital role in the development o...
research
05/31/2023

A Geometric Perspective on Diffusion Models

Recent years have witnessed significant progress in developing efficient...
research
06/15/2020

Inner Ensemble Nets

We introduce Inner Ensemble Networks (IENs) which reduce the variance wi...
research
10/24/2018

Distilling with Performance Enhanced Students

The task of accelerating large neural networks on general purpose hardwa...

Please sign up or login with your details

Forgot password? Click here to reset