Flexible Ensemble Research Articles

The Development of Flexible Ensemble in Thailand

Flexible ensemble is the arrangement that allows the selection of instruments for the instrumentation in order to overcome the problem of a small wind orchestra that cannot be performed with a combination of instruments according to the score for a standard big wind orchestra, or in the case of a wind orchestra where the exact number of musicians or types of instruments cannot be determined. Flexible ensemble is therefore another innovation that helps the teaching and performing management of a wind orchestra to be smooth and efficient, both in managing musicians for rehearsals and music performances. The purpose of this article is to provide the reader's understanding of flexible ensemble and to encourage their use and appreciation of its significance in order to produce musical literary compositions in the form of flexible bands with numerous variations in the future.

A Deep Ensemble Approach for Long-Term Traffic Flow Prediction

In the last 50 years, with the growth of cities and increase in the number of vehicles and mobility, traffic has become troublesome. As a result, traffic flow prediction started to attract attention as an important research area. However, despite the extensive literature, traffic flow prediction still remains as an open research problem, specifically for long-term traffic flow prediction. Compared to the models developed for short-term traffic flow prediction, the number of models developed for long-term traffic flow prediction is very few. Based on this shortcoming, in this study, we focus on long-term traffic flow prediction and propose a novel deep ensemble model (DEM). In order to build this ensemble model, first, we developed a convolutional neural network (CNN), a long short-term memory (LSTM) network and a gated recurrent unit (GRU) network as deep learning models, which formed the base learners. In the next step, we combine the output of these models according to their individual forecasting success. We use another deep learning model to determine the success of the individual models. Our proposed model is a flexible ensemble prediction model that can be updated based on traffic data. To evaluate the performance of the proposed model, we use a publicly available dataset. Experimental results show that the developed DEM model has a mean square error of 0.06 and a mean absolute error of 0.15 for single-step prediction; it shows that achieves a mean square error of 0.25 and a mean absolute error of 0.32 for multi-step prediction. We compared our proposed model with many models in different categories; individual deep learning models (i.e., LSTM, CNN, GRU), selected traditional machine learning models (i.e., linear regression, decision tree regression, k-nearest-neighbors regression) and other ensemble models such as random-forest regression. These results also support the claim that ensemble learning models perform better than individual models.

Application of the Improved K-Nearest Neighbor-Based Multi-Model Ensemble Method for Runoff Prediction

Hydrological forecasting plays a crucial role in mitigating flood risks and managing water resources. Data-driven hydrological models demonstrate exceptional fitting capabilities and adaptability. Recognizing the limitations of single-model forecasting, this study introduces an innovative approach known as the Improved K-Nearest Neighbor Multi-Model Ensemble (IKNN-MME) method to enhance the runoff prediction. IKNN-MME dynamically adjusts model weights based on the similarity of historical data, acknowledging the influence of different training data features on localized predictions. By combining an enhanced K-Nearest Neighbor (KNN) algorithm with adaptive weighting, it offers a more powerful and flexible ensemble. This study evaluates the performance of the IKNN-MME method across four basins in the United States and compares it to other multi-model ensemble methods and benchmark models. The results underscore its outstanding performance and adaptability, offering a promising avenue for improving runoff forecasting.

Neural-SEIR: A flexible data-driven framework for precise prediction of epidemic disease.

Accurately modeling and predicting epidemic diseases is crucial to prevent disease transmission and reduce mortality. Due to various unpredictable factors, including population migration, vaccination, control efforts, and seasonal fluctuations, traditional epidemic models that rely on prior knowledge of virus transmission mechanisms may not be sufficient to forecast complex epidemics like coronavirus disease 2019(COVID-19). The application of traditional epidemiological models such as susceptible-exposed-infectious-recovered (SEIR) may face difficulties in accurately predicting such complex epidemics. Data-driven prediction approaches lack the ability to generalize and exhibit low accuracy on small datasets due to their reliance on large amounts of data without incorporating prior knowledge. To overcome this limitation, we introduce a flexible ensemble data-driven framework (Neural-SEIR) that "neuralizes" the SEIR model by approximating the core parameters through neural networks while preserving the propagation structure of SEIR. Neural-SEIR employs long short-term memory (LSTM) neural network to capture complex correlation features, exponential smoothing (ES) to model seasonal information, and prior knowledge from SEIR. By incorporating SEIR parameters into the neural network structure, Neural-SEIR leverages prior knowledge while updating parameters with real-world data. Our experimental results demonstrate that Neural-SEIR outperforms traditional machine learning and epidemiological models, achieving high prediction accuracy and efficiency in forecasting epidemic diseases.

Development of a Flexible Ensemble Classification System for Microscopy

An abstract is not available for this content so a preview has been provided. As you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

A Flexible Ensemble Algorithm for Big Data Cleaning of PMUs

With an increasing application of Phase Measurement Units in the smart grid, it is becoming inevitable for PMUs to operate in severe conditions, which results in outliers and missing data. However, conventional techniques take excessive time to clean outliers and fill missing data due to lacking support from a big data platform. In this paper, a flexible ensemble algorithm is proposed to implement a precise and scalable data clean by the existing big data platform “Apache Spark.” In the proposed scheme, an ensemble model based on a soft voting approach utilizes principal component analysis in conjunction with the K-means, Gaussian mixture model, and isolation forest technique to detect outliers. The proposed scheme uses a gradient boosting decision tree for each extracted feature of PMUs for the data filling process after detecting outliers. The test results demonstrate that the proposed model achieves high accuracy and recall by comparing simulated and real-world Phase measurement unit data using the local outlier factor algorithm and Density-Based Spatial Clustering of Application with Noise (DBSCAN). The mean absolute error, root mean square error and R2-score criteria are used to validate the proposed method’s data filling results against contemporary techniques such as decision tree and linear regression algorithms.

Prediction of Protein Complex Structure Using Surface-Induced Dissociation and Cryo-Electron Microscopy.

A variety of techniques involving the use of mass spectrometry (MS) have been developed to obtain structural information on proteins and protein complexes. One example of these techniques, surface-induced dissociation (SID), has been used to study the oligomeric state and connectivity of protein complexes. Recently, we demonstrated that appearance energies (AE) could be extracted from SID experiments and that they correlate with structural features of specific protein-protein interfaces. While SID AE provides some structural information, the AE data alone are not sufficient to determine the structures of the complexes. For this reason, we sought to supplement the data with computational modeling, through protein-protein docking. In a previous study, we demonstrated that the scoring of structures generated from protein-protein docking could be improved with the inclusion of SID data; however, this work relied on knowledge of the correct tertiary structure and only built full complexes for a few cases. Here, we performed docking using input structures that require less prior knowledge, using homology models, unbound crystal structures, and bound+perturbed crystal structures. Using flexible ensemble docking (to build primarily subcomplexes from an ensemble of backbone structures), the RMSD100 of all (15/15) predicted structures using the combined Rosetta, cryo-electron microscopy (cryo-EM), and SID score was less than 4 Å, compared to only 7/15 without SID and cryo-EM. Symmetric docking (which used symmetry to build full complexes) resulted in predicted structures with RMSD100 less than 4 Å for 14/15 cases with experimental data, compared to only 5/15 without SID and cryo-EM. Finally, we also developed a confidence metric for which all (26/26) proteins flagged as high confidence were accurately predicted.

Artificial intelligence and healthcare: Forecasting of medical bookings through multi-source time-series fusion

Nowadays, Artificial intelligence (AI), combined with the digitalization of healthcare, can lead to substantial improvements in Patient Care, Disease Management, Hospital Administration, and supply chain effectiveness. Among predictive analytics tools, time series forecasting represents a central task to support healthcare management in terms of bookings and medical services predictions. In this context, the development of flexible frameworks to provide robust and reliable predictions became a central point in this healthcare innovation process. This paper presents and discusses a multi-source time series fusion and forecasting framework relying on Deep Learning. By combining weather, air-quality and medical bookings time series through a feature compression stage which preserves temporal patterns, the prediction is provided through a flexible ensemble technique based on machine learning models and a hybrid neural network. The proposed system is able to predict the number of bookings related to a specific medical examination for a 7-days horizon period. To assess the proposed approach’s effectiveness, we rely on time series extracted from a real dataset of administrative e-health records provided by the Campania Region health department, in Italy.

Big Archive-Assisted Ensemble of Many-Objective Evolutionary Algorithms

Multiobjective evolutionary algorithms (MOEAs) have witnessed prosperity in solving many-objective optimization problems (MaOPs) over the past three decades. Unfortunately, no one single MOEA equipped with given parameter settings, mating-variation operator, and environmental selection mechanism is suitable for obtaining a set of solutions with excellent convergence and diversity for various types of MaOPs. The reality is that different MOEAs show great differences in handling certain types of MaOPs. Aiming at these characteristics, this paper proposes a flexible ensemble framework, namely, ASES, which is highly scalable for embedding any number of MOEAs to promote their advantages. To alleviate the undesirable phenomenon that some promising solutions are discarded during the evolution process, a big archive that number of contained solutions be far larger than population size is integrated into this ensemble framework to record large-scale nondominated solutions, and also an efficient maintenance strategy is developed to update the archive. Furthermore, the knowledge coming from updating archive is exploited to guide the evolutionary process for different MOEAs, allocating limited computational resources for efficient algorithms. A large number of numerical experimental studies demonstrated superior performance of the proposed ASES. Among 52 test instances, the ASES performs better than all the six baseline algorithms on at least half of the test instances with respect to both metrics hypervolume and inverted generational distance.

Flexible iterative ensemble smoother for calibration of perfect and imperfect models

Iterative ensemble smoothers have been widely used for calibrating simulators of various physical systems due to the relatively low computational cost and the parallel nature of the algorithm. However, iterative ensemble smoothers have been designed for perfect models under the main assumption that the specified physical models and subsequent discretized mathematical models have the capability to model the reality accurately. While significant efforts are usually made to ensure the accuracy of the mathematical model, it is widely known that the physical models are only an approximation of reality. These approximations commonly introduce some type of model error which is generally unknown and when the models are calibrated, the effects of the model errors could be smeared by adjusting the model parameters to match historical observations. This results in a bias estimated parameters and as a consequence might result in predictions with questionable quality. In this paper, we formulate a flexible iterative ensemble smoother, which can be used to calibrate imperfect models where model errors cannot be neglected. We base our method on the ensemble smoother with multiple data assimilation (ES-MDA) as it is one of the most widely used iterative ensemble smoothing techniques. In the proposed algorithm, the residual (data mismatch) is split into two parts. One part is used to derive the parameter update and the second part is used to represent the model error. The proposed method is quite general and relaxes many of the assumptions commonly introduced in the literature. We observe that the proposed algorithm has the capability to reduce the effect of model bias by capturing the unknown model errors, thus improving the quality of the estimated parameters and prediction capacity of imperfect physical models.

BEES: Bayesian Ensemble Estimation from SAS

Many biomolecular complexes exist in a flexible ensemble of states in solution that is necessary to perform their biological function. Small-angle scattering (SAS) measurements are a popular method for characterizing these flexible molecules because of their relative ease of use and their ability to simultaneously probe the full ensemble of states. However, SAS data is typically low dimensional and difficult to interpret without the assistance of additional structural models. In theory, experimental SAS curves can be reconstituted from a linear combination of theoretical models, although this procedure carries a significant risk of overfitting the inherently low-dimensional SAS data. Previously, we developed a Bayesian-based method for fitting ensembles of model structures to experimental SAS data that rigorously avoids overfitting. However, we have found that these methods can be difficult to incorporate into typical SAS modeling workflows, especially for users that are not experts in computational modeling. To this end, we present the Bayesian Ensemble Estimation from SAS (BEES) program. Two forks of BEES are available, the primary one existing as a module for the SASSIE web server and a developmental version that is a stand-alone Python program. BEES allows users to exhaustively sample ensemble models constructed from a library of theoretical states and to interactively analyze and compare each model’s performance. The fitting routine also allows for secondary data sets to be supplied, thereby simultaneously fitting models to both SAS data as well as orthogonal information. The flexible ensemble of K63-linked ubiquitin trimers is presented as an example of BEES’ capabilities.

Predicting the influence of viral message for VM campaign on Weibo

Abstract Viral marketing (VM) method has the benefits of rapid dissemination and is easy to reach the “hard-to-get” audience, consequently, it has irritated the interest of academics and entrepreneurs. On social network sites, the prerequisite for successfully launching an effective VM campaign is to accurately predict the influence of viral message (IVM) under various seeding strategies that initiate and promote the campaign. However, in current prediction models of IVM, the effect of seeds on viral messages dissemination is ideally assumed to be constant and the corresponding seeding strategies are ignored. In this study, an IVM prediction model is proposed by taking the seeds’ influence with respect to infecting their followers into consideration. The attention features of seeding strategies, which get users to be attentive to a viral message, are identified and the flexible ensemble model (FEM) is developed to estimate seed’s influence under various seeding strategies. FEM evaluates the generalization ability of prediction models by generalization performance index and adaptively selects the appropriate model with the highest similarity to other models that perform well in generalization using rough similarity index. The proposed models are assessed with the real VM cases on Weibo and the experiment results show that FEM can offset the performance variability of traditional data mining models in different conversion volume samples and has strong robustness and adaptability.

Short-Term Forecasting for Energy Consumption through Stacking Heterogeneous Ensemble Learning Model

In the real-life, time-series data comprise a complicated pattern, hence it may be challenging to increase prediction accuracy rates by using machine learning and conventional statistical methods as single learners. This research outlines and investigates the Stacking Multi-Learning Ensemble (SMLE) model for time series prediction problem over various horizons with a focus on the forecasts accuracy, directions hit-rate, and the average growth rate of total oil demand. This investigation presents a flexible ensemble framework in light of blend heterogeneous models for demonstrating and forecasting nonlinear time series. The proposed SMLE model combines support vector regression (SVR), backpropagation neural network (BPNN), and linear regression (LR) learners, the ensemble architecture consists of four phases: generation, pruning, integration, and ensemble prediction task. We have conducted an empirical study to evaluate and compare the performance of SMLE using Global Oil Consumption (GOC). Thus, the assessment of the proposed model was conducted at single and multistep horizon prediction using unique benchmark techniques. The final results reveal that the proposed SMLE model outperforms all the other benchmark methods listed in this study at various levels such as error rate, similarity, and directional accuracy by 0.74%, 0.020%, and 91.24%, respectively. Therefore, this study demonstrates that the ensemble model is an extremely encouraging methodology for complex time series forecasting.

Estimating model‐error covariances in nonlinear state‐space models using Kalman smoothing and the expectation–maximization algorithm

Specification and tuning of errors from dynamical models are important issues in data assimilation. In this work, we propose an iterative expectation–maximization (EM) algorithm to estimate the model‐error covariances using classical extended and ensemble versions of the Kalman smoother. We show that, for additive model errors, the estimate of the error covariance converges. We also investigate other forms of model error, such as parametric or multiplicative errors. We show that additive Gaussian model error is able to compensate for non‐additive sources of error in the algorithms we propose. We also demonstrate the limitations of the extended version of the algorithm and recommend the use of the more robust and flexible ensemble version. This article is a proof of concept of the methodology with the Lorenz‐63 attractor. We developed an open‐source Python library to enable future users to apply the algorithm to their own nonlinear dynamical models.

Structure and calcium-binding studies of calmodulin-like domain of human non-muscle α-actinin-1

The activity of several cytosolic proteins critically depends on the concentration of calcium ions. One important intracellular calcium-sensing protein is α-actinin-1, the major actin crosslinking protein in focal adhesions and stress fibers. The actin crosslinking activity of α-actinin-1 has been proposed to be negatively regulated by calcium, but the underlying molecular mechanisms are poorly understood. To address this, we determined the first high-resolution NMR structure of its functional calmodulin-like domain (CaMD) in calcium-bound and calcium-free form. These structures reveal that in the absence of calcium, CaMD displays a conformationally flexible ensemble that undergoes a structural change upon calcium binding, leading to limited rotation of the N- and C-terminal lobes around the connecting linker and consequent stabilization of the calcium-loaded structure. Mutagenesis experiments, coupled with mass-spectrometry and isothermal calorimetry data designed to validate the calcium binding stoichiometry and binding site, showed that human non-muscle α-actinin-1 binds a single calcium ion within the N-terminal lobe. Finally, based on our structural data and analogy with other α-actinins, we provide a structural model of regulation of the actin crosslinking activity of α-actinin-1 where calcium induced structural stabilisation causes fastening of the juxtaposed actin binding domain, leading to impaired capacity to crosslink actin.

Variable importance and prediction methods for longitudinal problems with missing variables.

We present prediction and variable importance (VIM) methods for longitudinal data sets containing continuous and binary exposures subject to missingness. We demonstrate the use of these methods for prognosis of medical outcomes of severe trauma patients, a field in which current medical practice involves rules of thumb and scoring methods that only use a few variables and ignore the dynamic and high-dimensional nature of trauma recovery. Well-principled prediction and VIM methods can provide a tool to make care decisions informed by the high-dimensional patient’s physiological and clinical history. Our VIM parameters are analogous to slope coefficients in adjusted regressions, but are not dependent on a specific statistical model, nor require a certain functional form of the prediction regression to be estimated. In addition, they can be causally interpreted under causal and statistical assumptions as the expected outcome under time-specific clinical interventions, related to changes in the mean of the outcome if each individual experiences a specified change in the variable (keeping other variables in the model fixed). Better yet, the targeted MLE used is doubly robust and locally efficient. Because the proposed VIM does not constrain the prediction model fit, we use a very flexible ensemble learner (the SuperLearner), which returns a linear combination of a list of user-given algorithms. Not only is such a prediction algorithm intuitive appealing, it has theoretical justification as being asymptotically equivalent to the oracle selector. The results of the analysis show effects whose size and significance would have been not been found using a parametric approach (such as stepwise regression or LASSO). In addition, the procedure is even more compelling as the predictor on which it is based showed significant improvements in cross-validated fit, for instance area under the curve (AUC) for a receiver-operator curve (ROC). Thus, given that 1) our VIM applies to any model fitting procedure, 2) under assumptions has meaningful clinical (causal) interpretations and 3) has asymptotic (influence-curve) based robust inference, it provides a compelling alternative to existing methods for estimating variable importance in high-dimensional clinical (or other) data.

SEABED: Small molEcule activity scanner weB servicE baseD.

Motivation: The SEABED web server integrates a variety of docking and QSAR techniques in a user-friendly environment. SEABED goes beyond the basic docking and QSAR web tools and implements extended functionalities like receptor preparation, library editing, flexible ensemble docking, hybrid docking/QSAR experiments or virtual screening on protein mutants. SEABED is not a monolithic workflow tool but Software as a Service platform. Availability and implementation: SEABED is a free web server available athttp://www.bsc.es/SEABED. No registration is required. Contact: ramon.goni@bsc.es Supplementary information: Supplementary data are available atBioinformatics online.

Compensatory Flux Changes within an Endocytic Trafficking Network Maintain Thermal Robustness of Notch Signaling

SummaryDevelopmental signaling is remarkably robust to environmental variation, including temperature. For example, in ectothermic animals such as Drosophila, Notch signaling is maintained within functional limits across a wide temperature range. We combine experimental and computational approaches to show that temperature compensation of Notch signaling is achieved by an unexpected variety of endocytic-dependent routes to Notch activation which, when superimposed on ligand-induced activation, act as a robustness module. Thermal compensation arises through an altered balance of fluxes within competing trafficking routes, coupled with temperature-dependent ubiquitination of Notch. This flexible ensemble of trafficking routes supports Notch signaling at low temperature but can be switched to restrain Notch signaling at high temperature and thus compensates for the inherent temperature sensitivity of ligand-induced activation. The outcome is to extend the physiological range over which normal development can occur. Similar mechanisms may provide thermal robustness for other developmental signals.

The informatics philharmonic

I present ongoing work in developing a system that accompanies a live musician in a classical concerto-type setting, providing a flexible ensemble the follows the soloist in real-time and adapts to the soloist's interpretation through rehearsal. An accompanist must hear the soloist. The program models hearing through a hidden Markov model that can accurately and reliably parse highly complex audio in both offline and online fashion. The probabilistic formulation allows the program to navigate the latency/accuracy tradeoff in online following, so that onset detections occur with greater latency (and greater latency) when local ambiguities arise. For music with a sense of pulse, coordination between parts must be achieved by anticipating future evolution. The program develops a probabilistic model for musical timing, a Bayesian Belief Network, that allows the program to anticipate where future note onsets will occur, and to achieve better prediction using rehearsal data. The talk will include a live demonstration of the system on a staple from the violin concerto repertoire, as well as applications to more forward-looking interactions between soloist and computer controlled instruments.

Conformational behavior of peptides containing residues of 3-azetidinesulfonic (3AzeS) and 4-piperidinemethanesulfonic (4PiMS) acids

The conformational behavior of four model peptides containing residues of 3-azetidinesulfonic (3AzeS) and 4-piperidinemethane sulfonic (4PiMS) acids was studied in both a crystalline state and in solution using X-ray, NMR, and IR experiments. It was found that in the crystalline state, both of the models di- and tripeptides studied adopted extended conformations and demonstrated considerable conformational flexibility. In solution, it is likely that a flexible ensemble of conformations is adopted, including extended structures and more compact ones without persistent hydrogen bonds. One of the most interesting features of the peptides was the axial chirality observed due to the slow rotation around the amide bond formed by the endocyclic nitrogen atoms of the non-chiral 3AzeS and 4PiMS residues. It was shown for one of the derivatives that the configuration of the chiral axis had an impact on the conformation of the neighboring amino acid residue.