Tuning Parameter Values Research Articles

An efficient algorithm for estimating brain covariance networks.

Often derived from partial correlations or many pairwise analyses, covariance networks represent the inter-relationships among regions and can reveal important topological structures in brain measures from healthy and pathological subjects. However both approaches are not consistent network estimators and are sensitive to the value of the tuning parameters. Here, we propose a consistent covariance network estimator by maximising the network likelihood (MNL) which is robust to the tuning parameter. We validate the consistency of our algorithm theoretically and via a simulation study, and contrast these results against two well-known approaches: the graphical LASSO (gLASSO) and Pearson pairwise correlations (PPC) over a range of tuning parameters. The MNL algorithm had a specificity equal to and greater than 0.94 for all sample sizes in the simulation study, and the sensitivity was shown to increase as the sample size increased. The gLASSO and PPC demonstrated a specificity-sensitivity trade-off over a range of values of tuning parameters highlighting the discrepancy in the results for misspecified values. Application of the MNL algorithm to the case study data showed a loss of connections between healthy and impaired groups, and improved ability to identify between lobe connectivity in contrast to gLASSO networks. In this work, we propose the MNL algorithm as an effective approach to find covariance brain networks, which can inform the organisational features in brain-wide analyses, particularly for large sample sizes.

Pareto-Based Multi-objective Optimization for Fractional Order $$\hbox {PI}^{\lambda }$$ PI λ Speed Control of Induction Motor by Using Elman Neural Network

This paper presents Pareto-based multi-objective optimization for speed control of induction motor with fractional order proportional integral ( $$\hbox {FOPI}^{\lambda }$$ ) controller. The aim of this study is to find optimum values of tuning parameters of $$\hbox {FOPI}^{\lambda }$$ by using Elman neural network (ENN) and Pareto-based multi-objective optimization. In this context, proportional gain $$K_{\mathrm{p}}$$ , integral gain $$K_{\mathrm{i}}$$ and the order of fractional integral $$\lambda $$ are selected as tuning parameters while settling time $${T}_{\mathrm{s}}$$ and overshoot $${M}_{\mathrm{o}}$$ are chosen as objective functions. Firstly, experiments have been carried out to obtain training and test data. Then, ENN has been trained to construct mathematical model which is necessary for multi-objective optimization. In the next step, accuracy and reliability of ENN model are tested by using test data taken from experimental set-up. Finally, Pareto-based multi-objective optimization method has been used to find the optimum values of tuning parameters that minimize both $${T}_{\mathrm{s}}$$ and $${M}_{\mathrm{o}}$$ values. The different three conditions of the Pareto solution set are applied to the experimental set-up to verify the effectiveness of the proposed method. Results show that ENN is well modelled for induction motor and Pareto solution is an effective method to find optimal values of controller coefficients according to desired $${T}_{\mathrm{s}}$$ and $${M}_{\mathrm{o}}$$ values.

Sequential adaptive elastic net approach for single-snapshot source localization

This paper proposes efficient algorithms for accurate recovery of direction-of-arrivals (DoAs) of sources from single-snapshot measurements using compressed beamforming (CBF). In CBF, the conventional sensor array signal model is cast as an underdetermined complex-valued linear regression model and sparse signal recovery methods are used for solving the DoA finding problem. A complex-valued pathwise weighted elastic net (c-PW-WEN) algorithm is developed that finds solutions at the knots of penalty parameter values over a path (or grid) of elastic net (EN) tuning parameter values. c-PW-WEN also computes least absolute shrinkage and selection operator (LASSO) or weighted LASSO in its path. A sequential adaptive EN (SAEN) method is then proposed that is based on c-PW-WEN algorithm with adaptive weights that depend on previous solution. Extensive simulation studies illustrate that SAEN improves the probability of exact recovery of true support compared to conventional sparse signal recovery approaches such as LASSO, EN, or orthogonal matching pursuit in several challenging multiple target scenarios. The effectiveness of SAEN is more pronounced in the presence of high mutual coherence.

Consensus Classification Using Non-Optimized Classifiers.

Classifying samples into categories is a common problem in analytical chemistry and other fields. Classification is usually based on only one method, but numerous classifiers are available with some being complex, such as neural networks, and others are simple, such as k nearest neighbors. Regardless, most classification schemes require optimization of one or more tuning parameters for best classification accuracy, sensitivity, and specificity. A process not requiring exact selection of tuning parameter values would be useful. To improve classification, several ensemble approaches have been used in past work to combine classification results from multiple optimized single classifiers. The collection of classifications for a particular sample are then combined by a fusion process such as majority vote to form the final classification. Presented in this Article is a method to classify a sample by combining multiple classification methods without specifically classifying the sample by each method, that is, the classification methods are not optimized. The approach is demonstrated on three analytical data sets. The first is a beer authentication set with samples measured on five instruments, allowing fusion of multiple instruments by three ways. The second data set is composed of textile samples from three classes based on Raman spectra. This data set is used to demonstrate the ability to classify simultaneously with different data preprocessing strategies, thereby reducing the need to determine the ideal preprocessing method, a common prerequisite for accurate classification. The third data set contains three wine cultivars for three classes measured at 13 unique chemical and physical variables. In all cases, fusion of nonoptimized classifiers improves classification. Also presented are atypical uses of Procrustes analysis and extended inverted signal correction (EISC) for distinguishing sample similarities to respective classes.

Sensitivity analysis via Karhunen–Loève expansion of a random field model: Estimation of Sobol’ indices and experimental design

We use the Karhunen–Loève expansion of a random-field model to construct a tensorised Bayesian linear model from which Sobol’ sensitivity indices can be estimated straightforwardly. The method combines the advantages of models built from families of orthonormal functions, which facilitate computations, and Gaussian-process models, which offer a lot of flexibility. The posterior distribution of the indices can be derived, and its normal approximation can be used to design experiments especially adapted to their estimation. Implementation details are provided, and values of tuning parameters are indicated that yield precise estimation from a small number of function evaluations. Several illustrative examples are included that show the good performance of the method, in particular in comparison with estimation based on polynomial chaos expansion.

Immigration: How Does It Impact the Wages of Native Workers?

I have worried about the talk, in recent times, that immigrants hurt the wages of native workers in the host nation. If so, that is not a good outcome. Why should native workers lose out to immigrants? To come to terms with my worry, I began to experiment with a classic dataset on immigrants and native worker wages, which was assembled about 15 years ago. At that time, this dataset appeared to signal that there exists an immigrant effect independent of skill set (e.g., education) that reduces the wages of native workers. Specifically, the indication was that the proportion of immigrant workers (p) is a standalone exogenous variable, among others, in a linear regression model that can explain native worker wages without including, in the model, the interaction of p with skill set. In my experimentation, I find that p cannot be a standalone variable in the model and at the same time: 1) be competitive in terms of cross validation predicted residual error sum of squares; and 2) be immune to the shocks from ridge regression, for small values of the ridge tuning parameter. My experiments indicate that immigration differentially impacts native worker wages across skill sets. In particular, for a 10% increase in immigrants, the average wage of native workers is impacted as follows: a) there is about a 5% drop (in native worker wages) for High school dropouts; b) about an 8% drop for High school graduates; c) about a 2% rise for those with some college education; and d) about a 6% rise for College graduates. This impact is, generally, monotonic in skill. Based on my experiments, I am led to cultivate the view that the overall (or net) impact of immigrants on native worker wages is based on their skill set shares. Several researchers have found little overall impact of immigrants on native worker wages; and, based on my view, it is not at all clear to me that there is anything unreasonable about their finding. That being said, Governments should nudge, and facilitate, native workers to acquire skills beyond those a High school education can provide; and ensure that labor market tests are correctly applied, especially in circumstances where it becomes quite difficult to price the deployment of a lower level skill like mopping the floor or driving a truck. Disciplined immigration will not harm the fabric of a nation. Economic inequality will.

Optimisation of grid connected hybrid photovoltaic–wind–battery system using model predictive control design

This study explores optimisation of the hybrid power system in the smart grid framework, in conjunction with the model predictive control (MPC) design. This study also creates a strategy that can maximise the use of renewable energy, e.g. photovoltaic, the wind turbine with battery storage and minimise the utilisation of the utility grid for electricity usage in the industry. This is devised by modelling a discrete state-space model of the hybrid power system for a given industry application. The system design is implemented within a real-time electricity pricing environment that is integrated with renewable energy to optimally meet the demand according to a specific performance of the consumer. The emphasis of this approach is on its capacity to supply optimal power to the demand side by selecting the appropriate source; and its robustness against uncertainties. The results show that MPC design for hybrid power system not only optimises the energy flow but also improves the overall process of energy management. It was also observed that the optimal solution minimises the delay cost of energy demand from the utility grid according to a given reference from the consumer for the specified tuning parameter values of the performance index.

Autonomous robotic exploration using a utility function based on Rényi’s general theory of entropy

In this paper we present a novel information-theoretic utility function for selecting actions in a robot-based autonomous exploration task. The robot’s goal in an autonomous exploration task is to create a complete, high-quality map of an unknown environment as quickly as possible. This implicitly requires the robot to maintain an accurate estimate of its pose as it explores both unknown and previously observed terrain in order to correctly incorporate new information into the map. Our utility function simultaneously considers uncertainty in both the robot pose and the map in a novel way and is computed as the difference between the Shannon and the Renyi entropy of the current distribution over maps. Renyi’s entropy is a family of functions parameterized by a scalar, with Shannon’s entropy being the limit as this scalar approaches unity. We link the value of this scalar parameter to the predicted future uncertainty in the robot’s pose after taking an exploratory action. This effectively decreases the expected information gain of the action, with higher uncertainty in the robot’s pose leading to a smaller expected information gain. Our objective function allows the robot to automatically trade off between exploration and exploitation in a way that does not require manually tuning parameter values, a significant advantage over many competing methods that only use Shannon’s definition of entropy. We use simulated experiments to compare the performance of our proposed utility function to these state-of-the-art utility functions. We show that robots that use our proposed utility function generate maps with less uncertainty and fewer visible artifacts and that the robots have less uncertainty in their pose during exploration. Finally, we demonstrate that a real-world robot using our proposed utility function is able to successfully create a high-quality map of an indoor office environment.

Correction to Consensus Outlier Detection Using Sum of Ranking Differences of Common and New Outlier Measures Without Tuning Parameter Selections.

Sample outlier detection is imperative before calculating a multivariate calibration model. Outliers, especially in high-dimensional space, can be difficult to detect. The outlier measures Hotelling’s t-squared, Q-residuals, and Studentized residuals are standard in analytical chemistry with spectroscopic data. However, these and other merits are tuning parameter dependent and sensitive to the outlier themselves, i.e., the measures are susceptible to swamping and masking. Additionally, different samples are also often flagged as outliers depending on the outlier measure used. Sum of ranking differences (SRD) is a new generic fusion tool that can simultaneously evaluate multiple outlier measures across windows of tuning parameter values thereby simplifying outlier detection and providing improved detection. Presented in this paper is SRD to detect multiple outliers despite the effects of masking and swamping. Both spectral (x-outlier) and analyte (y-outlier) outliers can be detected separately or in tandem...

Understanding the dynamics of terrorism events with multiple-discipline datasets and machine learning approach.

Terror events can cause profound consequences for the whole society. Finding out the regularity of terrorist attacks has important meaning for the global counter-terrorism strategy. In the present study, we demonstrate a novel method using relatively popular and robust machine learning methods to simulate the risk of terrorist attacks at a global scale based on multiple resources, long time series and globally distributed datasets. Historical data from 1970 to 2015 was adopted to train and evaluate machine learning models. The model performed fairly well in predicting the places where terror events might occur in 2015, with a success rate of 96.6%. Moreover, it is noteworthy that the model with optimized tuning parameter values successfully predicted 2,037 terrorism event locations where a terrorist attack had never happened before.

Consensus Outlier Detection Using Sum of Ranking Differences of Common and New Outlier Measures Without Tuning Parameter Selections.

Sample outlier detection is imperative before calculating a multivariate calibration model. Outliers, especially in high-dimensional space, can be difficult to detect. The outlier measures Hotelling's t-squared, Q-residuals, and Studentized residuals are standard in analytical chemistry with spectroscopic data. However, these and other merits are tuning parameter dependent and sensitive to the outlier themselves, i.e., the measures are susceptible to swamping and masking. Additionally, different samples are also often flagged as outliers depending on the outlier measure used. Sum of ranking differences (SRD) is a new generic fusion tool that can simultaneously evaluate multiple outlier measures across windows of tuning parameter values thereby simplifying outlier detection and providing improved detection. Presented in this paper is SRD to detect multiple outliers despite the effects of masking and swamping. Both spectral (x-outlier) and analyte (y-outlier) outliers can be detected separately or in tandem with SRD using respective merits. Unique to SRD are fusion verification processes to confirm samples flagged as outliers. The SRD process also allows for sample masking checks. Presented, and used by SRD, are several new outlier detection measures. These measures include atypical uses of Procrustes analysis and extended inverted signal correction (EISC). The methodologies are demonstrated on two near-infrared (NIR) data sets.

Active Incremental Recognition of Human Activities in a Streaming Context

Recognising human activities from streaming sources poses unique challenges to learning algorithms. Predictive models need to be scalable, incrementally trainable, and must remain bounded in size even when the data stream is arbitrarily long. In order to achieve high accuracy even in complex and dynamic environments methods should be also nonparametric, i.e., their structure should adapt in response to the incoming data. Furthermore, as tuning is problematic in a streaming setting, suitable approaches should be parameterless (as initially tuned parameter values may not prove optimal for future streams). Here, we present an approach to the recognition of human actions from streaming data which meets all these requirements by: (1) incrementally learning a model which adaptively covers the feature space with simple and local classifiers; (2) employing an active learning strategy to reduce annotation requests; (3) achieving good accuracy within a fixed model size. Although in this work we focus on human activity recognition, our approach is completely independent from the feature extraction and can deal with any supervised matrix (set of feature vectors). Hence, it can be adapted to a wide range of applications (e.g., speech recognition, image classification, object recognition, pose recognition, and image matching). Extensive experiments on standard benchmarks show that our approach is competitive with state-of-the-art non-incremental methods, while outperforming the existing active incremental baselines.

Sample‐wise spectral multivariate calibration desensitized to new artifacts relative to the calibration data using a residual penalty

Calibration maintenance is an important aspect of multivariate calibration. With spectral measurements, the goal of calibration maintenance involves sustaining the predictability of a primary calibration model in new secondary conditions. Among the many methodologies, penalty‐based Tikhonov regularization variants have been successful by sample augmenting primary calibration data with a matrix of just a few secondary samples as well as operating with an additional sparse penalty to include wavelength selection. Studied in this paper is a new sample‐wise (local) Tikhonov regularization–based penalty calibration approach. Penalized is a diagonal matrix with the residual vector (relative to the primary calibration space) of the new secondary sample. Thus, the same full calibration set is used for each new sample. Changing for each secondary sample is the corresponding sample‐wise residual vector on the penalized diagonal matrix. The intent of the presented approach is to form sample‐wise regression vectors desensitized to characteristics of the new sample not present in the primary calibration set. The more distinct the secondary conditions are relative to the primary conditions, the more unsuccessful this local model updating becomes. Proposed is a sample‐wise outlier mechanism to discern when the residual penalty can or cannot be used to form a useful updated model. The residual penalty modeling and outlier detection processes require tuning parameter optimizations. A fusion approach is used to automatically select tuning parameter values. Simulated and near‐infrared data are evaluated, demonstrating the applicability of the method.

Designing a Channel Access Mechanism for Wireless Sensor Network

Although there are various Medium Access Control (MAC) protocols proposed for Wireless Sensor Network (WSN), there is no protocol accepted as a standard specific to it. This paper deals with completing the design of our previously proposed MAC for WSN by proposing a channel access mechanism (CAM). The CAM is based on developing a backoff mechanism which mainly differentiates nodes’ backoffs depending on their different identification numbers, and it employs a performance tuning parameter for reaching a required performance objective. The probability distribution of the backoff period is constructed and Markov chain modeling is used to analyze and evaluate the CAM against the IEEE802.15.4 slotted CSMA/CA based on single- and multihop communication with respect to the reliability, the average delay, the power consumption, and the throughput. The analysis reveals that the required performance of CAM against the IEEE slotted CSMA/CA can be obtained by choosing the maximum backoff stages number and the tuning parameter value and that CAM performs better than the IEEE with larger nodes number. The multihop scenario results in a good end-to-end performance of CAM with respect to the reliability and delay becomes better with lengthier paths at the expense of increasing the energy consumption.

Load Frequency Control of an Isolated Micro Grid Using Fuzzy Adaptive Model Predictive Control

This paper presents a fuzzy adaptive model predictive approach for load frequency control of an isolated micro grid. A generalized state space model of a typical isolated micro grid having controllable and uncontrollable generating power sources is derived, and the same has been utilized to predict the future output and control inputs for the micro grid frequency control. A centralized model predictive control (MPC) is implemented with a single input multi-output system model based on the controllable distributed energy resources in the micro grid. The parameter-driven MPC is made adaptable by dynamically adjusting its parameter $R_{w}$ using fuzzy controller. The proposed fuzzy MPC employs a rule-based fuzzy controller to fuzzify the tuning parameter $R_{w}$ present in the cost function of MPC, which plays an important role in minimizing the frequency deviation in the system. The closed loop system response obtained by the proposed fuzzy MPC has been found faster and adaptable for different scenarios in the system. The effectiveness has been evaluated with performance index integral time square error (ITSE) value and has been compared with MPC with constant tuning parameter value also with the proportional–integral controller response. Thus, the efficacy of using proposed fuzzy MPC in secondary load frequency control has been validated thereof.

The influence of fouling on the dynamic behavior of PID-controlled heat exchangers

The influence of fouling build-up on the dynamic behavior of PID-controlled shell-and-tube heat exchangers was studied. A mathematical model was proposed in which heat exchanger is considered as an array of cells where the exchange of heat is modelled using lumped-parameter approach. The model was implemented employing Simulink/MATLAB software and validated using real-life data records acquired during 3years of operation of a heat exchanger network coupled with a crude distillation unit. Using the validated dynamic model, a simulation study was carried out on selected PID-controlled exchangers operated in the mentioned network. For different periods of uninterrupted exchanger operation, that is, at different values of the thermal resistance of fouling, suitable controller tuning parameters were studied. The results proved that at changed thermal resistance of fouling, the values of tuning parameters should be changed too.

Tuning Parameters of Pid Controllers for the Operation of Heat Exchangers Under Fouling Conditions

The present paper is an extension of the work presented by authors at SDEWES2015 conference, where changes in the dynamic behaviour of shell-and-tube heat exchangers induced by fouling build-up were studied. A mathematical model was proposed of transient heat exchange with the influence of thermal resistance of fouling taken into account. Next a mathematical model of the entire heat exchanger was implemented and simulations were carried out using Simulink in MATLAB environment. The model was validated using data records acquired during 3 years of operation of a heat exchanger network coupled with a Crude Distillation Unit (CDU). Using the validated dynamic model, one can simulate the behaviour of a heat exchanger together with its PID control loop and investigate the influence of fouling build-up on specific indexes of quality of operation. A simulation study was carried out on selected PID-controlled exchangers operated in a heat exchanger network coupled with a CDU rated 220 kg/s crude oil. For different periods of uninterrupted exchanger operation, that is, at different values of the thermal resistance of fouling, tuning parameters were determined using the Ziegler-Nichols method. It was found that the optimal values of tuning parameters changed when the thermal resistance of fouling was changed. However, oscillations remained in the characteristics of some controlled exchangers indicating that the new values of tuning parameters determined by the simple Ziegler–Nichols method were ineffective. In the present work, to improve the quality indices of the exchangers control under fouling conditions, a more advanced method of the optimization of controller tuning is presented using Signal Constraint block in MATLAB/SIMULINK.

Automatic support vector data description

Event handlers have wide range of applications such as medical assistant systems and fire suppression systems. These systems try to provide accurate responses based on the least information. Support vector data description (SVDD) is one of the appropriate tools for such detections, which should handle lack of information. Therefore, many efforts have been done to improve SVDD. Unfortunately, the existing descriptors suffer from weak data characteristic in sparse data sets and their tuning parameters are organized improperly. These issues cause reduction of accuracy in event handlers when they are faced with data shortage. Therefore, we propose automatic support vector data description (ASVDD) based on both validation degree, which is originated from fuzzy rough set to discover data characteristic, and assigning effective values for tuning parameters by chaotic bat algorithm. To evaluate the performance of ASVDD, several experiments have been conducted on various data sets of UCI repository. The experimental results demonstrate superiority of the proposed method over state-of-the-art ones in terms of classification accuracy and AUC. In order to prove meaningful distinction between the accuracy results of the proposed method and the leading-edge ones, the Wilcoxon statistical test has been conducted.

On Stabilization of Quantized Sampled-Data Neural-Network-Based Control Systems.

This paper investigates the problem of stabilization of sampled-data neural-network-based systems with state quantization. Different with previous works, the communication limitation of state quantization is considered for the first time. More specifically, it is assumed that the sampled state measurements from sensor to the controller are quantized via a quantizer. To reduce conservativeness, a novel piecewise Lyapunov-Krasovskii functional (LKF) is constructed by introducing a line-integral type Lyapunov function and some useful terms that take full advantage of the available information about the actual sampling pattern. Based on the new LKF, much less conservative stabilization conditions are derived to obtain the maximal sampling period and the minimal guaranteed cost control performance. The desired quantized sampled-data three-layer fully connected feedforward neural-network-based controllers are designed by a linear matrix inequality approach. A search algorithm is given to find the optimal values of tuning parameters. The effectiveness and advantage of proposed method are demonstrated by the numerical simulation of an inverted pendulum.

A Causal, Data-driven Approach to Modeling the Kepler Data

Astronomical observations are affected by several kinds of noise, each with its own causal source; there is photon noise, stochastic source variability, and residuals coming from imperfect calibration of the detector or telescope. The precision of NASA Kepler photometry for exoplanet science—the most precise photometric measurements of stars ever made—appears to be limited by unknown or untracked variations in spacecraft pointing and temperature, and unmodeled stellar variability. Here, we present the causal pixel model (CPM) for Kepler data, a data-driven model intended to capture variability but preserve transit signals. The CPM works at the pixel level so that it can capture very fine-grained information about the variation of the spacecraft. The CPM models the systematic effects in the time series of a pixel using the pixels of many other stars and the assumption that any shared signal in these causally disconnected light curves is caused by instrumental effects. In addition, we use the target star’s future and past (autoregression). By appropriately separating, for each data point, the data into training and test sets, we ensure that information about any transit will be perfectly isolated from the model. The method has four tuning parameters—the number of predictor stars or pixels, the autoregressive window size, and two L2-regularization amplitudes for model components, which we set by cross-validation. We determine values for tuning parameters that works well for most of the stars and apply the method to a corresponding set of target stars. We find that CPM can consistently produce low-noise light curves. In this paper, we demonstrate that pixel-level de-trending is possible while retaining transit signals, and we think that methods like CPM are generally applicable and might be useful for K2, TESS, etc., where the data are not clean postage stamps like Kepler.