High Probability Regions Research Articles

Implicit Sampling for Path Integral Control, Monte Carlo Localization, and SLAM

Implicit sampling is a recently developed variationally enhanced sampling method that guides its samples to regions of high probability, so that each sample carries information. Implicit sampling may thus improve the performance of algorithms that rely on Monte Carlo (MC) methods. Here the applicability and usefulness of implicit sampling for improving the performance of MC methods in estimation and control is explored, and implicit sampling based algorithms for stochastic optimal control, stochastic localization, and simultaneous localization and mapping (SLAM) are presented. The algorithms are tested in numerical experiments where it is found that fewer samples are required if implicit sampling is used, and that the overall runtimes of the algorithms are reduced.

Enhancing sensitivity and specificity in rare cell capture microdevices with dielectrophoresis.

The capture and subsequent analysis of rare cells, such as circulating tumor cells from a peripheral blood sample, has the potential to advance our understanding and treatment of a wide range of diseases. There is a particular need for high purity (i.e., high specificity) techniques to isolate these cells, reducing the time and cost required for single-cell genetic analyses by decreasing the number of contaminating cells analyzed. Previous work has shown that antibody-based immunocapture can be combined with dielectrophoresis (DEP) to differentially isolate cancer cells from leukocytes in a characterization device. Here, we build on that work by developing numerical simulations that identify microfluidic obstacle array geometries where DEP-immunocapture can be used to maximize the capture of target rare cells, while minimizing the capture of contaminating cells. We consider geometries with electrodes offset from the array and parallel to the fluid flow, maximizing the magnitude of the resulting electric field at the obstacles' leading and trailing edges, and minimizing it at the obstacles' shoulders. This configuration attracts cells with a positive DEP (pDEP) response to the leading edge, where the shear stress is low and residence time is long, resulting in a high capture probability; although these cells are also repelled from the shoulder region, the high local fluid velocity at the shoulder minimizes the impact on the overall transport and capture. Likewise, cells undergoing negative DEP (nDEP) are repelled from regions of high capture probability and attracted to regions where capture is unlikely. These simulations predict that DEP can be used to reduce the probability of capturing contaminating peripheral blood mononuclear cells (using nDEP) from 0.16 to 0.01 while simultaneously increasing the capture of several pancreatic cancer cell lines from 0.03-0.10 to 0.14-0.55, laying the groundwork for the experimental study of hybrid DEP-immunocapture obstacle array microdevices.

基于BP神经网络的延边地区渤海国遗址预测研究

本文选择同遗址分布有关的高程、坡度、坡向、距离河流的距离、距离村屯的距离等因素值的数据集为样本，利用BP神经网络建立延边地区渤海国遗址预测模型，并对预测结果分析。结果表明：预测模型的预测准确率达88.7%，高概率区面积占研究区域的17.6%。预测结果高概率区高程集中在海拔270~570 m高程区间，坡度集中在0˚~3˚区间，坡向集中在平地和东北方向，高概率区在河流缓冲区上，主要分布在0~1000 m缓冲区间，在道路缓冲区上主要分布在0~1500 m缓冲区间，在村屯缓冲区上主要分布在500~1500 m缓冲区间。研究结果为今后发掘新的遗址和保护现存的遗址方面提供一定的科学依据。 In this study, we selected data sets as the sample which was related to the distribution of the site; the factors included height, gradient, slope aspect, the distance from the river, and the distance from village etc. This is based on a back-propagation (BP) neural net work to establish forecasting model to analyze the predicting results on Bohai kingdom site in Yanbian Korean Autonomous Prefecture. The results suggest that the accuracy of the prediction model gets to 88.7%, the high probability region of the whole study area is 17.6%. The elevation value of model prediction results of the high probability area concentrates in 270 - 570 meters, the slope is concentrated in 0 - 3 degree, the slope direction is concentrated in the flat ground and the Northeast. The high proba-bility region of the river buffer is distributed from 0 to 1000 meters. The site areas of road buffer are distributed within 0 - 1500 meters and the site areas of the village buffer are distributed within 500 - 1500 meters. These results provide the scientific foundations for excavation and protection of the sites.

Modeling analysis on the silica glass synthesis in a hydrogen diffusion flame

Silica glass ingot synthesis by flame hydrolysis deposition (FHD) is an important approach to obtain high-purity synthetic silica glass with high refractive index homogeneity. Using the precursor of silicon tetrachloride (SiCl4), the silica ingot (SiO2) can be synthesized in the oxy-hydrogen diffusion flame through a set of kinetic reactions and subsequent cooling. The homogeneity of the synthetic silica glass is influenced by the temperature/diameter equality of silica droplets and the residue of radicals (e.g. OH) in the synthetic silica. In this study, the mixing system of oxy-hydrogen diffusion flame and silica droplets in a furnace were modeled by a Euler–Lagrange multi-phase model, where the interphase exchange of mass, momentum and energy between the gas phase molecules and liquid phase silica droplets during the formation and transportation of silica droplets were included. The silica droplets were tracked by a Lagrangian formulation that includes the discrete phase inertia, hydrodynamic drag, the force of gravity and the dispersion due to turbulent eddies. The molecular gas phase reactions are described by a simple kinetic mechanism involving the major species during the formation of molecular SiO2, with kinetic and thermodynamic parameters taken from the literature when available.The effect of initial equivalence ratio on the flame structure was analyzed in the study, where the high-temperature region is uniform over the glass ingot and the OH residue is limited at unity equivalence ratio. The probability distributions of silica droplet diameter and temperature were analyzed by employing a developed droplet growth model, where the condensation rate is controlled by the in situ SiO2 vapor concentration and the local flow-condition-determined mass transfer rate. The maximum diameter of silica droplets in the furnace is 1.39×10−4m, and the percentage of silica droplets decreases with the increasing of diameter. Two high probability regions were observed for the droplet temperature distribution, the temperature below 1500K accounts for 40% of the total droplet number, and the temperature range between 3500 and 4500K accounts for 58%. The droplet diameter on the ingot cap distributes in the range from 4×10−5 to 9×10−5m with approximately probability of 85%. The distribution of droplet temperature on the ingot cap is much uniform, with more than half (67%) of the droplets has the temperature between 3100 and 3200K.

Cooperative Search of Multiple Unknown Transient Radio Sources Using Multiple Paired Mobile Robots

We develop a localization method to enable a team of mobile robots to search for multiple unknown transient radio sources. Because of signal source anonymity, short transmission durations, and dynamic transmission patterns, robots cannot treat the radio sources as continuous radio beacons. Moreover, robots do not know the source transmission power and have limited sensing ranges. To cope with these challenges, we pair up robots and develop a cooperative sensing model using signal strength ratios from the paired robots. We formally prove that the joint conditional posterior probability of source locations for the m-robot team can be obtained by combining the pairwise joint posterior probabilities, which can be derived from signal strength ratios. Moreover, we propose a pairwise ridge walking algorithm (PRWA) to coordinate the robot pairs based on the clustering of high-probability regions and the minimization of local Shannon entropy. We have implemented and validated the algorithm under both the hardware-driven simulation and physical experiments. Experimental results show that the PRWA-based localization scheme consistently outperforms the other four heuristics.

Atomistic insight into orthoborate-based ionic liquids: force field development and evaluation.

We have developed an all-atomistic force field for a new class of halogen-free chelated orthoborate-phosphonium ionic liquids. The force field is based on an AMBER framework with determination of force field parameters for phosphorus and boron atoms, as well as refinement of several available parameters. The bond and angle force constants were adjusted to fit vibration frequency data derived from both experimental measurements and ab initio calculations. The force field parameters for several dihedral angles were obtained by fitting torsion energy profiles deduced from ab initio calculations. To validate the proposed force field parameters, atomistic simulations were performed for 12 ionic liquids consisting of tetraalkylphosphonium cations and chelated orthoborate anions. The predicted densities for neat ionic liquids and the [P6,6,6,14][BOB] sample, with a water content of approximately 2.3-2.5 wt %, are in excellent agreement with available experimental data. The potential energy components of 12 ionic liquids were discussed in detail. The radial distribution functions and spatial distribution functions were analyzed and visualized to probe the microscopic ionic structures of these ionic liquids. There are mainly four high-probability regions of chelated orthoborate anions distributed around tetraalkylphosphonium cations in the first solvation shell, and such probability distribution functions are strongly influenced by the size of anions.

A Square Root Unscented FastSLAM With Improved Proposal Distribution and Resampling

An improved square root unscented fast simultaneous localization and mapping (FastSLAM) is proposed in this paper. The proposed method propagates and updates the square root of the state covariance directly in Cholesky decomposition form. Since the choice of the proposal distribution and that of the resampling method are the most critical issues to ensure the performance of the algorithm, its optimization is considered by improving the sampling and resampling steps. For this purpose, particle swarm optimization (PSO) is used to optimize the proposal distribution. PSO causes the particle set to tend to the high probability region of the posterior before the weights are updated; thereby, the impoverishment of particles can be overcome. Moreover, a new resampling algorithm is presented to improve the resampling step. The new resampling algorithm can conquer the defects of the resampling algorithm and solve the degeneracy and sample impoverishment problem simultaneously. Compared to unscented FastSLAM (UFastSLAM), the proposed algorithm can maintain the diversity of particles and consequently avoid inconsistency for longer time periods, and furthermore, it can improve the estimation accuracy compared to UFastSLAM. These advantages are verified by simulations and experimental tests for benchmark environments.

Improving counterflow detection in dense crowds with scene features

This paper addresses the problem of detecting counterflow motion in videos of highly dense crowds. We focus on improving the detection performance by identifying scene features – that is, features on motionless background surfaces. We propose a three-way classifier to differentiate counterflow from normal flow, simultaneously identifying scene features based on statistics of low-level feature point tracks. By monitoring scene features, we can reduce the likelihood that moving features’ point tracks mix with scene feature point tracks, as well as detect and discard frames with periodic jitter. We also construct a Scene Feature Heat Map, which reflects the space-varying probability that object trajectories might mix with scene features. When an object trajectory nears a high-probability region of this map, we switch to a more time-consuming and robust joint Lucas–Kanade tracking algorithm to improve performance. We evaluate the algorithms with extensive experiments on several datasets, including almost three weeks of data from an airport surveillance camera network. The experiments demonstrate the feasibility of the proposed algorithms and their significant improvements for counterflow detection.

Segment selective dynamic histogram equalization for brightness preserving contrast enhancement of images

Histogram equalization (HE), is a simple and less complex technique for image contrast enhancement, but it fails to preserve their brightness and natural appearance. It usually introduces visual artifacts due to saturation and un-even expansion of intensities. This paper proposes a novel modification of HE method, which decomposes the histogram of the input image into multiple segments utilizing median values as thresholds. Prior to equalization process, segments having high probability regions are identified and allocated with new dynamic range, while other segments remain unaltered. In order to reduce the algorithmic complexity, the HE is applied over selected segments instead of complete histogram. Finally combination of equalized and un-equalized segments are normalized. Simulation results show that for wide-range of test images, the proposed method enhances contrast while preserving the brightness and natural appearance of the images.

Comparison of Seismic Design Schemes for Suspension Bridge in Mountain Region with Strong Earthquake

The Longjiang Bridge is a kilometer level suspension bridge supported by reinforced concrete towers. The towers will be founded on pile group foundation, and the project area is in a mountain region of high earthquake probability. Two seismic schemes were designed for the bridge, and two computational bridge models in full size were established, extensive studies were performed to capture the seismic response of different schemes for the earthquakes. The results show that towers composed of two curved quadrilateral hollow concrete columns are adopted for the bridge.

An adaptive sparse-grid high-order stochastic collocation method for Bayesian inference in groundwater reactive transport modeling

[1] Bayesian analysis has become vital to uncertainty quantification in groundwater modeling, but its application has been hindered by the computational cost associated with numerous model executions required by exploring the posterior probability density function (PPDF) of model parameters. This is particularly the case when the PPDF is estimated using Markov Chain Monte Carlo (MCMC) sampling. In this study, a new approach is developed to improve the computational efficiency of Bayesian inference by constructing a surrogate of the PPDF, using an adaptive sparse-grid high-order stochastic collocation (aSG-hSC) method. Unlike previous works using first-order hierarchical basis, this paper utilizes a compactly supported higher-order hierarchical basis to construct the surrogate system, resulting in a significant reduction in the number of required model executions. In addition, using the hierarchical surplus as an error indicator allows locally adaptive refinement of sparse grids in the parameter space, which further improves computational efficiency. To efficiently build the surrogate system for the PPDF with multiple significant modes, optimization techniques are used to identify the modes, for which high-probability regions are defined and components of the aSG-hSC approximation are constructed. After the surrogate is determined, the PPDF can be evaluated by sampling the surrogate system directly without model execution, resulting in improved efficiency of the surrogate-based MCMC compared with conventional MCMC. The developed method is evaluated using two synthetic groundwater reactive transport models. The first example involves coupled linear reactions and demonstrates the accuracy of our high-order hierarchical basis approach in approximating high-dimensional posteriori distribution. The second example is highly nonlinear because of the reactions of uranium surface complexation, and demonstrates how the iterative aSG-hSC method is able to capture multimodal and non-Gaussian features of PPDF caused by model nonlinearity. Both experiments show that aSG-hSC is an effective and efficient tool for Bayesian inference.

Implicit Particle Methods and Their Connection with Variational Data Assimilation

AbstractThe implicit particle filter is a sequential Monte Carlo method for data assimilation that guides the particles to the high-probability regions via a sequence of steps that includes minimizations. A new and more general derivation of this approach is presented and the method is extended to particle smoothing as well as to data assimilation for perfect models. Minimizations required by implicit particle methods are shown to be similar to those that one encounters in variational data assimilation, and the connection of implicit particle methods with variational data assimilation is explored. In particular, it is argued that existing variational codes can be converted into implicit particle methods at a low additional cost, often yielding better estimates that are also equipped with quantitative measures of the uncertainty. A detailed example is presented.

Implicit Sampling, with Application to Data Assimilation

There are many computational tasks in which it is necessary to sample a given probability density function (or pdf for short), i.e., to use a computer to construct a sequence of independent random vectors x i (i=1,2,…), whose histogram converges to the given pdf. This can be difficult because the sample space can be huge, and more importantly, because the portion of the space where the density is significant, can be very small, so that one may miss it by an ill-designed sampling scheme. Indeed, Markov-chain Monte Carlo, the most widely used sampling scheme, can be thought of as a search algorithm, where one starts at an arbitrary point and one advances step-by-step towards the high probability region of the space. This can be expensive, in particular because one is typically interested in independent samples, while the chain has a memory. The authors present an alternative, in which samples are found by solving an algebraic equation with a random right-hand side rather than by following a chain; each sample is independent of the previous samples. The construction is explained in the context of numerical integration, and it is then applied to data assimilation.

Spatial probability cuing and right hemisphere damage

In this experiment we studied statistical learning, inter-trial priming, and visual attention. We assessed healthy controls and right brain damaged (RBD) patients with and without neglect, on a simple visual discrimination task designed to measure priming effects and probability learning. All participants showed a preserved priming effect for item color. Contrary to healthy controls and RBD participants without neglect, RBD participants with neglect did not show positional priming and both RBD groups learned the underlying spatial probability distribution of target locations to a lesser degree. To see if the latter deficiency could be improved, we tested a patient with long standing chronic spatial neglect on three separate days and observed improved identification times for left sided, high probability, targets. In summary, we found preservation of priming per se in people with spatial neglect. However, this was only clearly demonstrable for color priming and not for positional priming. Associated with this impairment was a difficulty in learning the overall statistical structure of target locations. In a patient with severe persistent neglect we were able to demonstrate that the deficit in statistical learning was not absolute, as this subject improved his identification times for targets appearing in high probability regions of the test display.

Constrained MSSM favoring new territories: The impact of new LHC limits and a 125 GeV Higgs boson

We present an updated and extended global analysis of the constrained MSSM (CMSSM) taking into account new limits on supersymmetry from $\ensuremath{\sim}5/\mathrm{fb}$ data sets at the LHC. In particular, in the case of the razor limit obtained by the CMS Collaboration we simulate detector efficiency for the experimental analysis and derive an approximate but accurate likelihood function. We discuss the impact on the global fit of a possible Higgs boson with mass near 125 GeV, as implied by recent data, and of a new improved limit on $\mathrm{BR}({\mathrm{B}}_{\mathrm{s}}\ensuremath{\rightarrow}{\ensuremath{\mu}}^{+}{\ensuremath{\mu}}^{\ensuremath{-}})$. We identify high posterior probability regions of the CMSSM parameters as the stau-coannihilation and the $A$-funnel region, with the importance of the latter now being much larger due to the combined effect of the above three LHC results and of dark matter relic density. We also find that the focus point region is now disfavored. Ensuing implications for superpartner masses favor even larger values than before, and even lower ranges for dark matter spin-independent cross section, ${\ensuremath{\sigma}}_{p}^{\mathrm{SI}}\ensuremath{\lesssim}{10}^{\ensuremath{-}9}\text{ }\text{ }\mathrm{pb}$. We also find that relatively minor variations in applying experimental constraints can induce a large shift in the location of the best-fit point. This puts into question the robustness of applying the usual ${\ensuremath{\chi}}^{2}$ approach to the CMSSM. We discuss the goodness-of-fit and find that, while it is difficult to calculate a $p$-value, the $(g\ensuremath{-}2{)}_{\ensuremath{\mu}}$ constraint makes, nevertheless, the overall fit of the CMSSM poor. We consider a scan without this constraint, and we allow $\ensuremath{\mu}$ to be either positive or negative. We find that the global fit improves enormously for both signs of $\ensuremath{\mu}$, with a slight preference for $\ensuremath{\mu}<0$ caused by a better fit to $\mathrm{BR}(\mathrm{b}\ensuremath{\rightarrow}s\ensuremath{\gamma})$ and $\mathrm{BR}({\mathrm{B}}_{\mathrm{s}}\ensuremath{\rightarrow}{\ensuremath{\mu}}^{+}{\ensuremath{\mu}}^{\ensuremath{-}})$.

An Efficient Method for Structural Reliability Analysis

An efficient reliability analysis method has been proposed in the paper, which based on stochastic response surface method. The key advantage of the stochastic response surface method is that the collocation points are selected for minimizing the mean square error, and from high probability regions, thus leading to fewer function evaluations for high accuracy. Compared with Monte Carlo method, the proposed technique can be more efficiency while achieve comparative accuracy.

Dual estimation of partially observed nonlinear structural systems: A particle filter approach

Dual estimation consists of tracking the whole state of partially observed systems, and simultaneously estimating unknown model parameters. In case of nonlinearly evolving systems, standard filtering procedures may provide unreliable model calibrations, either because of estimates affected by bias or due to diverging filter response. In this paper, we propose a particle filter (PF) wherein particles, i.e. system realizations evolving in a stochastic frame, are first sampled from the current probability density function of the system and then moved towards the region of high probability by an extended Kalman filter. We show that the proposed filter works much better than a standard PF, in terms of accuracy of the estimates and of computing time.

Simultaneous Localization of Multiple Unknown and Transient Radio Sources Using a Mobile Robot

We report system and algorithm developments that utilize a single mobile robot to simultaneously localize multiple unknown transient radio sources. Because of signal source anonymity, short transmission durations, and dynamic transmission patterns, the robot cannot treat the radio sources as continuous radio beacons. To deal with this challenging localization problem, we model the radio source behaviors using a novel spatiotemporal probability occupancy grid that captures transient characteristics of radio transmissions and tracks posterior probability distributions of radio sources. As a Monte Carlo method, a ridge walking motion planning algorithm is proposed to enable the robot to efficiently traverse the high-probability regions to accelerate the convergence of the posterior probability distribution. We also formally show that the time to find a radio source is insensitive to the number of radio sources, and hence, our algorithm has great scalability. We have implemented the algorithms and extensively tested them in comparison with two heuristic methods: a random walk and a fixed-route patrol. The localization time of our algorithms is consistently shorter than that of the two heuristic methods.

Efficient fully nonlinear data assimilation for geophysical fluid dynamics

A potential problem with Ensemble Kalman Filter is the implicit Gaussian assumption at analysis times. Here we explore the performance of a recently proposed fully nonlinear particle filter on a high-dimensional but simplified ocean model, in which the Gaussian assumption is not made. The model simulates the evolution of the vorticity field in time, described by the barotropic vorticity equation, in a highly nonlinear flow regime. While common knowledge is that particle filters are inefficient and need large numbers of model runs to avoid degeneracy, the newly developed particle filter needs only of the order of 10–100 particles on large scale problems. The crucial new ingredient is that the proposal density can not only be used to ensure all particles end up in high-probability regions of state space as defined by the observations, but also to ensure that most of the particles have similar weights. Using identical twin experiments we found that the ensemble mean follows the truth reliably, and the difference from the truth is captured by the ensemble spread. A rank histogram is used to show that the truth run is indistinguishable from any of the particles, showing statistical consistency of the method.

Correspondence Priors for Binocular Image Data via Canonical Correlation Analysis

Event Abstract Back to Event Correspondence Priors for Binocular Image Data via Canonical Correlation Analysis Christian Conrad1* and Rudolf Mester1 1 J.W.G University Frankfurt, Visual Sensorics and Information Processing Group, Germany In this work, we study unsupervised learning of correspondences relations in binocular video streams. This is useful for low level vision tasks in stereo vision or motion estimation but also for analysis of fMRI data. Still an open question is, how correspondence relations evolve and are represented in the human brain. Correspondence estimation is often based on the principle of identifying corresponding pixels or patches and appears in many forms: in stereo vision pixel correspondences among a pair of images taken at the same point in time serve to determine a depth map [LucasKanade1981]. In motion estimation, pixel correspondences among consecutive images are sought [HornSchunk1981]. Most spatial feature approaches to correspondence estimation are based on the prototypical detection and matching framework. Here, correspondences are typically not determined using the raw pixel representation but rather rich feature descriptors. In the past, feature detectors have often been designed based on statistical (and biological) principles [Schmid2000,Mikolajczyk2004]. Today, there is an increased interest in unsupervised learning of such features directly from data based on energy-models and probabilistic generative models [OlshausenField1997,Hyvaerinen2009]. Furthermore, unsupervised feature learning is not restricted to model the actual image content but can also be used to learn the relationship between pairs of images , where images may be related via a spatial transformation, depth map, or the optical flow [Susskind2011,Memisevic2012]. In contrast to probabilistic methods for unsupervised feature learning, often involving rather sophisticated machinery and optimization schemes, we present a sampling-free algorithm based on Canonical Correlation Analysis (CCA, [Hotelling1936]), and show how 'correspondence priors' can be determined in closed form. Specifically, given video streams of two views, our algorithm first determines pixel correspondences on a coarse scale via learning the inter-image transformation employing CCA. Subsequently it projects those correspondences to the original resolution. After learning, for each point in video channel A, regions of high probability containing the true correspondence are determined, thus forming correspondence priors. While CCA only allows us to learn the inter-image transformation implicitly, we show how to apply a learnt transformation to previously unseen data in a principled way. Correspondence priors are efficiently encoded using second order statistics and may then be plugged into probabilistic and energy based formulations of specific vision applications. In contrast popular probabilistic models, our algorithm can be applied in closed form, only involving QR decomposition or the SVD, thus making it especially suitable within real world applications. We experimentally verify the applicability of the approach in several real world scenarios where the binocular views may be subject to substantial spatial transformations. Acknowledgements We gratefully acknowledge funding by the German Federal Ministry of Education and Research (BMBF) in the project Bernstein Fokus Neurotechnologie -- Frankfurt Vision Initiative 01GQ0841. Keywords: CCA, Correspondence Estimation, Unsupervised Feature Learning Conference: Bernstein Conference 2012, Munich, Germany, 12 Sep - 14 Sep, 2012. Presentation Type: Poster Topic: Data analysis, machine learning, neuroinformatics Citation: Conrad C and Mester R (2012). Correspondence Priors for Binocular Image Data via Canonical Correlation Analysis. Front. Comput. Neurosci. Conference Abstract: Bernstein Conference 2012. doi: 10.3389/conf.fncom.2012.55.00124 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 11 May 2012; Published Online: 12 Sep 2012. * Correspondence: Mr. Christian Conrad, J.W.G University Frankfurt, Visual Sensorics and Information Processing Group, Frankfurt, Germany, christian.conrad@vsi.cs.uni-frankfurt.de Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers Christian Conrad Rudolf Mester Google Christian Conrad Rudolf Mester Google Scholar Christian Conrad Rudolf Mester PubMed Christian Conrad Rudolf Mester Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.