The Experts below are selected from a list of 11385 Experts worldwide ranked by ideXlab platform
Krzysztof Bartoszek - One of the best experts on this subject based on the ideXlab platform.
-
the phylogenetic Effective Sample Size and jumps
Mathematica Applicanda, 2018Co-Authors: Krzysztof BartoszekAbstract:The phylogenetic Effective Sample Size is a parameter that has as its goal the quantification of the amount of independent signal in a phylogenetically correlatedSample. It was studied for Brownian ...
-
phylogenetic Effective Sample Size
Journal of Theoretical Biology, 2016Co-Authors: Krzysztof BartoszekAbstract:In this paper I address the question—how large is a phylogenetic Sample? I propose a definition of a phylogenetic Effective Sample Size for Brownian motion and Ornstein-Uhlenbeck processes-the regr ...
-
phylogenetic Effective Sample Size
bioRxiv, 2015Co-Authors: Krzysztof BartoszekAbstract:In this paper I address the question - how large is a phylogenetic Sample? I propose a definition of a phylogenetic Effective Sample Size for Brownian motion and Ornstein-Uhlenbeck processes - the regression Effective Sample Size. I discuss how mutual information can be used to define an Effective Sample Size in the non-normal process case and compare these two definitions to an already present concept of Effective Sample Size (the mean Effective Sample Size). Through a simulation study I find that the AICc is robust if one corrects for the number of species or Effective number of species. Lastly I discuss how the concept of the phylogenetic Effective Sample Size can be useful for biodiversity quantification, identification of interesting clades and deciding on the importance of phylogenetic correlations
Ali Taylan Cemgil - One of the best experts on this subject based on the ideXlab platform.
-
EUSIPCO - Distributed particle filtering under real-time constraints
2017 25th European Signal Processing Conference (EUSIPCO), 2017Co-Authors: Alper Kamil Bozkurt, Ali Taylan CemgilAbstract:Particle filters are powerful methods for state estimation in nonlinear/non-Gaussian dynamical systems. However due to the heavy computational requirements, they may not satisfy the real-time constraints in many applications requiring a large number of particles. By means of distributed implementation, real-time particle filtering can be achieved. However, the resampling stage in particle filters requires particle interaction which causes communication overhead. In this work, we propose a distributed resampling algorithm based on Butterfly Resampling previously described in the literature. We describe three interaction schemes (i) the complete interaction, (ii) the pairwise interaction where the nodes are constrained to communicate in pairs and (iii) the partial pairwise interaction in which only one pair is allowed to communicate. The goal is to diminish the communication cost in exchange for negligible loss of Effective Sample Size. We conduct experiments on a cluster environment and compare our methods in terms of execution time, communication time and Effective Sample Size. We find that the sparse interaction schemes show better performance for distributed systems and they keep the Effective Sample Size nearly as high as the complete interaction scheme does.
-
Distributed particle filtering under real-time constraints
2017 25th European Signal Processing Conference (EUSIPCO), 2017Co-Authors: Alper Kamil Bozkurt, Ali Taylan CemgilAbstract:Particle filters are powerful methods for state estimation in nonlinear/non-Gaussian dynamical systems. However due to the heavy computational requirements, they may not satisfy the real-time constraints in many applications requiring a large number of particles. By means of distributed implementation, real-time particle filtering can be achieved. However, the resampling stage in particle filters requires particle interaction which causes communication overhead. In this work, we propose a distributed resampling algorithm based on Butterfly Resampling previously described in the literature. We describe three interaction schemes (i) the complete interaction, (ii) the pairwise interaction where the nodes are constrained to communicate in pairs and (iii) the partial pairwise interaction in which only one pair is allowed to communicate. The goal is to diminish the communication cost in exchange for negligible loss of Effective Sample Size. We conduct experiments on a cluster environment and compare our methods in terms of execution time, communication time and Effective Sample Size. We find that the sparse interaction schemes show better performance for distributed systems and they keep the Effective Sample Size nearly as high as the complete interaction scheme does.
Daniel M Zuckerman - One of the best experts on this subject based on the ideXlab platform.
-
automated sampling assessment for molecular simulations using the Effective Sample Size
Journal of Chemical Theory and Computation, 2010Co-Authors: Xin Zhang, Divesh Bhatt, Daniel M ZuckermanAbstract:To quantify the progress in the development of algorithms and force fields used in molecular simulations, a general method for the assessment of the sampling quality is needed. Statistical mechanics principles suggest the populations of physical states characterize equilibrium sampling in a fundamental way. We therefore develop an approach for analyzing the variances in state populations, which quantifies the degree of sampling in terms of the Effective Sample Size (ESS). The ESS estimates the number of statistically independent configurations contained in a simulated ensemble. The method is applicable to both traditional dynamics simulations as well as more modern (e.g., multicanonical) approaches. Our procedure is tested in a variety of systems from toy models to atomistic protein simulations. We also introduce a simple automated procedure to obtain approximate physical states from dynamic trajectories: this allows Sample-Size estimation in systems for which physical states are not known in advance.
-
automated sampling assessment for molecular simulations using the Effective Sample Size
arXiv: Computational Physics, 2010Co-Authors: Xin Zhang, Divesh Bhatt, Daniel M ZuckermanAbstract:To quantify the progress in development of algorithms and forcefields used in molecular simulations, a method for the assessment of the sampling quality is needed. We propose a general method to assess the sampling quality through the estimation of the number of independent Samples obtained from molecular simulations. This method is applicable to both dynamic and nondynamic methods and utilizes the variance in the populations of physical states to determine the ESS. We test the correctness and robustness of our procedure in a variety of systems--two-state toy model, all-atom butane, coarse-grained calmodulin, all-atom dileucine and Met-enkaphalin. We also introduce an automated procedure to obtain approximate physical states from dynamic trajectories: this procedure allows for Sample--Size estimation for systems for which physical states are not known in advance.
-
evaluating the Effective Sample Size of equilibrium molecular simulations using automatically approximated physical states
Biophysical Journal, 2009Co-Authors: Xin Q Zhang, Daniel M ZuckermanAbstract:In order to assess “convergence” in molecular simulations and to quantify the efficiency of competing algorithms, we need a reasonable and universally applicable estimate of the “Effective Sample Size,” N_eff. For equilibrium sampling, we suggest the most undamental definition of N_eff to be that number governing the variance in populations of physical states measured from multiple independent simulations. We demonstrate a simple automated procedure for approximating physical states and show that the resulting estimates for N_eff agree well with intuitive transition counts. A wide variety of biomolecular systems are successfully analyzed. Our approach can be applied to systems with unknown physical states and to modern non-dynamical algorithms, such as those based on the “exchange” mechanism.The necessary software for estimating N_eff will be freely available on our website.
-
on the structural convergence of biomolecular simulations by determination of the Effective Sample Size
Journal of Physical Chemistry B, 2007Co-Authors: Edward Lyman, Daniel M ZuckermanAbstract:Although atomistic simulations of proteins and other biological systems are approaching microsecond timescales, the quality of simulation trajectories has remained difficult to assess. Such assessment is critical not only for establishing the relevance of any individual simulation but also in the extremely active field of developing computational methods. Here we map the trajectory assessment problem onto a simple statistical calculation of the "Effective Sample Size", that is, the number of statistically independent configurations. The mapping is achieved by asking the question, "How much time must elapse between snapshots included in a Sample for that Sample to exhibit the statistical properties expected for independent and identically distributed configurations?" Our method is more general than autocorrelation methods in that it directly probes the configuration-space distribution without requiring a priori definition of configurational substates and without any fitting parameters. We show that the method is equally and directly applicable to toy models, peptides, and a 72-residue protein model. Variants of our approach can readily be applied to a wide range of physical and chemical systems.
F. Louzada - One of the best experts on this subject based on the ideXlab platform.
-
Effective Sample Size for importance sampling based on discrepancy measures
Signal Processing, 2017Co-Authors: Luca Martino, Victor Elvira, F. LouzadaAbstract:The Effective Sample Size (ESS) is an important measure of efficiency of Monte Carlo methods such as Markov Chain Monte Carlo (MCMC) and Importance Sampling (IS) techniques. In the IS context, an approximation ESS ^ of the theoretical ESS definition is widely applied, involving the inverse of the sum of the squares of the normalized importance weights. This formula, ESS ^ , has become an essential piece within Sequential Monte Carlo (SMC) methods, to assess the convenience of a resampling step. From another perspective, the expression ESS ^ is related to the Euclidean distance between the probability mass described by the normalized weights and the discrete uniform probability mass function (pmf). In this work, we derive other possible ESS functions based on different discrepancy measures between these two pmfs. Several examples are provided involving, for instance, the geometric mean of the weights, the discrete entropy (including the perplexity measure, already proposed in literature) and the Gini coefficient among others. We list five theoretical requirements which a generic ESS function should satisfy, allowing us to classify different ESS measures. We also compare the most promising ones by means of numerical simulations.
-
Alternative Effective Sample Size measures for importance sampling
2016 IEEE Statistical Signal Processing Workshop (SSP), 2016Co-Authors: L. Martino, V Elvira, F. LouzadaAbstract:The Effective Sample Size (ESS) is an important measure of efficiency in the Importance Sampling (IS) technique. A well-known approximation of the theoretical ESS definition, involving the inverse of the sum of the squares of the normalized importance weights, is widely applied in literature. This expression has become an essential piece within Sequential Monte Carlo (SMC) methods, using adaptive resampling procedures. In this work, first we show that this ESS approximation is related to the Euclidean distance between the probability mass function (pmf) described by the normalized weights and the uniform pmf. Then, we derive other possible ESS functions based on different discrepancy measures. In our study, we also include another ESS measure called perplexity, already proposed in literature, that is based on the discrete entropy of the normalized weights. We compare all of them by means of numerical simulations.
-
SSP - Alternative Effective Sample Size measures for importance sampling
2016 IEEE Statistical Signal Processing Workshop (SSP), 2016Co-Authors: Luca Martino, Victor Elvira, F. LouzadaAbstract:The Effective Sample Size (ESS) is an important measure of efficiency in the Importance Sampling (IS) technique. A well-known approximation of the theoretical ESS definition, involving the inverse of the sum of the squares of the normalized importance weights, is widely applied in literature. This expression has become an essential piece within Sequential Monte Carlo (SMC) methods, using adaptive resampling procedures. In this work, first we show that this ESS approximation is related to the Euclidean distance between the probability mass function (pmf) described by the normalized weights and the uniform pmf. Then, we derive other possible ESS functions based on different discrepancy measures. In our study, we also include another ESS measure called perplexity, already proposed in literature, that is based on the discrete entropy of the normalized weights. We compare all of them by means of numerical simulations.
-
Effective Sample Size for importance sampling based on discrepancy measures
viXra, 2016Co-Authors: Luca Martino, Victor Elvira, F. LouzadaAbstract:The Effective Sample Size (ESS) is an important measure of efficiency of Monte Carlo methods such as Markov Chain Monte Carlo (MCMC) and Importance Sampling (IS) techniques. In the IS context, an approximation $\widehat{ESS}$ of the theoretical ESS definition is widely applied, involving the inverse of the sum of the squares of the normalized importance weights. This formula, $\widehat{ESS}$, has become an essential piece within Sequential Monte Carlo (SMC) methods, to assess the convenience of a resampling step. From another perspective, the expression $\widehat{ESS}$ is related to the Euclidean distance between the probability mass described by the normalized weights and the discrete uniform probability mass function (pmf). In this work, we derive other possible ESS functions based on different discrepancy measures between these two pmfs. Several examples are provided involving, for instance, the geometric mean of the weights, the discrete entropy (including the {\it perplexity} measure, already proposed in literature) and the Gini coefficient among others. We list five theoretical requirements which a generic ESS function should satisfy, allowing us to classify different ESS measures. We also compare the most promising ones by means of numerical simulations.
Alper Kamil Bozkurt - One of the best experts on this subject based on the ideXlab platform.
-
EUSIPCO - Distributed particle filtering under real-time constraints
2017 25th European Signal Processing Conference (EUSIPCO), 2017Co-Authors: Alper Kamil Bozkurt, Ali Taylan CemgilAbstract:Particle filters are powerful methods for state estimation in nonlinear/non-Gaussian dynamical systems. However due to the heavy computational requirements, they may not satisfy the real-time constraints in many applications requiring a large number of particles. By means of distributed implementation, real-time particle filtering can be achieved. However, the resampling stage in particle filters requires particle interaction which causes communication overhead. In this work, we propose a distributed resampling algorithm based on Butterfly Resampling previously described in the literature. We describe three interaction schemes (i) the complete interaction, (ii) the pairwise interaction where the nodes are constrained to communicate in pairs and (iii) the partial pairwise interaction in which only one pair is allowed to communicate. The goal is to diminish the communication cost in exchange for negligible loss of Effective Sample Size. We conduct experiments on a cluster environment and compare our methods in terms of execution time, communication time and Effective Sample Size. We find that the sparse interaction schemes show better performance for distributed systems and they keep the Effective Sample Size nearly as high as the complete interaction scheme does.
-
Distributed particle filtering under real-time constraints
2017 25th European Signal Processing Conference (EUSIPCO), 2017Co-Authors: Alper Kamil Bozkurt, Ali Taylan CemgilAbstract:Particle filters are powerful methods for state estimation in nonlinear/non-Gaussian dynamical systems. However due to the heavy computational requirements, they may not satisfy the real-time constraints in many applications requiring a large number of particles. By means of distributed implementation, real-time particle filtering can be achieved. However, the resampling stage in particle filters requires particle interaction which causes communication overhead. In this work, we propose a distributed resampling algorithm based on Butterfly Resampling previously described in the literature. We describe three interaction schemes (i) the complete interaction, (ii) the pairwise interaction where the nodes are constrained to communicate in pairs and (iii) the partial pairwise interaction in which only one pair is allowed to communicate. The goal is to diminish the communication cost in exchange for negligible loss of Effective Sample Size. We conduct experiments on a cluster environment and compare our methods in terms of execution time, communication time and Effective Sample Size. We find that the sparse interaction schemes show better performance for distributed systems and they keep the Effective Sample Size nearly as high as the complete interaction scheme does.
