Leapfrogging and synoptic Leapfrogging: A new optimization approach

Rhinehart, R. Russell; Su, Ming; Manimegalai-Sridhar, Upasana

doi:10.1016/j.compchemeng.2012.02.011

Cited by 22 publications

(6 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the increased accuracy and complexity (i.e., number of parameters) of modern force fields, sophisticated high-dimensional optimization, multiobjective Pareto front, − and uncertainty quantification (UQ) methods − should play a key role in force field development. However, these methods are not tractable when molecular simulation is performed at each step of the algorithm, as this may necessitate O (10 2 to 10 6 ) simulations .…”

Section: Introductionmentioning

confidence: 99%

Configuration-Sampling-Based Surrogate Models for Rapid Parameterization of Non-Bonded Interactions

Messerly

Razavi

Shirts

2018

J. Chem. Theory Comput.

View full text Add to dashboard Cite

In this study, we present an approach for rapid force field parameterization and uncertainty quantification of the non-bonded interaction parameters for classical force fields. The accuracy of most thermophysical properties, and especially vapor-liquid equilibria (VLE), obtained from molecular simulation depends strongly on the non-bonded interactions. Traditionally, non-bonded interactions are parameterized to agree with macroscopic properties by performing large amounts of direct molecular simulation. Due to the computational cost of molecular simulation, surrogate models (i.e., efficient models that approximate direct molecular simulation results) are an essential tool for high-dimensional parameterization and uncertainty quantification of non-bonded interactions. The present study compares two different configuration-sampling-based surrogate models, namely, Multistate Bennett Acceptance Ratio (MBAR) and Pair Correlation Function Rescaling (PCFR). MBAR and PCFR are coupled with the Isothermal Isochoric (ITIC) thermodynamic integration method for estimating vapor-liquid saturation properties. We find that MBAR and PCFR are complementary in their roles. Specifically, PCFR is preferred when exploring distant regions of the parameter space while MBAR is better in the local domain.

show abstract

Section: Introductionmentioning

confidence: 99%

Configuration-Sampling-Based Surrogate Models for Rapid Parameterization of Non-Bonded Interactions

Messerly

Razavi

Shirts

2018

J. Chem. Theory Comput.

View full text Add to dashboard Cite

show abstract

“…For our purpose, we have employed the Bayesian regularization minimization algorithm which is one of the popular approaches to ANN training within the framework of the MATLAB nnstart toolbox and particularly useful for noisy data (Gençay and Qi, 2001). We remark that other training methods may also be used such as Lebenberg-Marquardt (Levenberg, 1944) and nature-inspired heuristic algorithms such as particle swarm (Eberhart and Kennedy, 1995), differential evolution (Storn and Price, 1997) or leapfrogging (Rhinehart et al, 2012). However, access to gradient information (through the use of a differentiable cost-function) heavily favors the use of gradient-based optimizers.…”

Section: Ann Trainingmentioning

confidence: 99%

An artificial neural network framework for reduced order modeling of transient flows

San

Maulik

Ahmed

2019

Communications in Nonlinear Science and Numerical Simulation

140

View full text Add to dashboard Cite

This paper proposes a supervised machine learning framework for the non-intrusive model order reduction of unsteady fluid flows to provide accurate predictions of non-stationary state variables when the control parameter values vary. Our approach utilizes a training process from full-order scale direct numerical simulation data projected on proper orthogonal decomposition (POD) modes to achieve an artificial neural network (ANN) model with reduced memory requirements. This data-driven ANN framework allows for a nonlinear time evolution of the modal coefficients without performing a Galerkin projection. Our POD-ANN framework can thus be considered an equation-free approach for latent space dynamics evolution of nonlinear transient systems and can be applied to a wide range of physical and engineering applications. Within this framework we introduce two architectures, namely sequential network (SN) and residual network (RN), to train the trajectory of modal coefficients. We perform a systematic analysis of the performance of the proposed reduced order modeling approaches on prediction of a nonlinear wave-propagation problem governed by the viscous Burgers equation, a simplified prototype setting for transient flows. We find that the POD-ANN-RN yields stable and accurate results for test problems assessed both within inside and outside of the database range and performs significantly better than the standard intrusive Galerkin projection model. Our results show that the proposed framework provides a non-intrusive alternative to the evolution of transient physics in a POD basis spanned space, and can be used as a robust predictive model order reduction tool for nonlinear dynamical systems.

show abstract

“…For example, the authors in [133] design a memetic algorithm for the maximum clique problem, i.e., an EA augmented with a local search. For illustration purposes, the rest of this section describes the outlines of the binary particle swarm optimization, an important instance of EA with a robust implementation, e.g., Leapfrogging [134]- [136]. A comparison of the performance of the different EAs can be found in [137].…”

Section: Randomized Solvers For Integer Problemsmentioning

confidence: 99%

A Tutorial on Clique Problems in Communications and Signal Processing

et al. 2020

View full text Add to dashboard Cite

Since its first use by Euler on the seven bridges of Königsberg problem, graph theory has shown excellent abilities in solving and unveiling the properties of multiple discrete optimization problems notably in communications and signal processing. The study of the structure of integer programs reveals equivalence with graph theory problems, which makes a large body of the literature readily available for solving and characterizing the complexity of these problems. This tutorial presents a framework for utilizing a particular graph theory problem, known as the clique problem, for solving communications and signal processing problems. In particular, the paper aims to illustrate the structural properties of problems that can be formulated as clique problems through multiple examples in communications and signal processing. To that end, the first part of the tutorial formulates and suggests various optimal and heuristic solutions for the clique problem and its variants. The tutorial, further, illustrates the use of the clique formulation and its variants through numerous contemporary examples in communications and signal processing, mainly in maximum access problem for non-orthogonal multiple access networks, throughput maximization using index coding and instantly decodable network coding, collision-free radio frequency identification networks, and resource allocation in cloud-radio access networks. The tutorial finally sheds light on the recent advances of such applications and discusses future research directions on the topics.

show abstract

Leapfrogging and synoptic Leapfrogging: A new optimization approach

Cited by 22 publications

References 30 publications

Configuration-Sampling-Based Surrogate Models for Rapid Parameterization of Non-Bonded Interactions

Configuration-Sampling-Based Surrogate Models for Rapid Parameterization of Non-Bonded Interactions

An artificial neural network framework for reduced order modeling of transient flows

A Tutorial on Clique Problems in Communications and Signal Processing

Contact Info

Product

Resources

About