Input Weights Research Articles

Illumination estimation via random vector functional link based on improved arithmetic optimization algorithm

AbstractIn order to solve the problem that the existing illumination estimation methods have low accuracy in image restoration, a random vector functional link illumination estimation algorithm based on whale optimization arithmetic algorithm is proposed in this article. First, in order to improve the optimization accuracy of arithmetic optimization algorithm, the whale optimization algorithm is applied to optimize the initial population of the arithmetic optimization algorithm. Then, we utilize the improved arithmetic optimization algorithm to improve the input weight and hidden layer offset of random vector functional link, and exclude the uncertainty caused by the random generation of random vector functional link's input weight and hidden layer bias. Finally, the proposed whale arithmetic optimization algorithm‐random vector functional link model is applied to SFU lab dataset and Gehler‐Shi dataset for illumination estimation. Compared with the other earlier illumination estimation algorithms, the proposed illumination estimation model in this article has better effect and higher stability. In addition, due to the diagonal mapping method for color correction and image restoration, the proposed illumination estimation algorithm can also get better results in image restoration.

Prediction model for cost data of a power transmission and transformation project based on Pearson correlation coefficient–IPSO–ELM

Abstract In view of the difficulty in predicting the cost data of power transmission and transformation projects at present, a method based on Pearson correlation coefficient–improved particle swarm optimization (IPSO)–extreme learning machine (ELM) is proposed. In this paper, the Pearson correlation coefficient is used to screen out the main influencing factors as the input-independent variables of the ELM algorithm and IPSO based on a ladder-structure coding method is used to optimize the number of hidden-layer nodes, input weights and bias values of the ELM. Therefore, the prediction model for the cost data of power transmission and transformation projects based on the Pearson correlation coefficient–IPSO–ELM algorithm is constructed. Through the analysis of calculation examples, it is proved that the prediction accuracy of the proposed method is higher than that of other algorithms, which verifies the effectiveness of the model.

Data multiplexed and hardware reused architecture for deep neural network accelerator

Despite many decades of research on high-performance Deep Neural Network (DNN) accelerators, their massive computational demand still requires resource-efficient, optimized and parallel architecture for computational acceleration. Contemporary hardware implementations of DNNs face the burden of excess area requirement due to resource-intensive elements such as multipliers and non-linear Activation Functions (AFs). This paper proposes DNN with reused hardware-costly AF by multiplexing data using shift-register. The on-chip quantized log2 based memory addressing with an optimized technique is used to access input features, weights, and biases. This way the external memory bandwidth requirement is reduced and dynamically adjusted for DNNs. Further, high-throughput and resource-efficient memory elements for sigmoid activation function are extracted using the Taylor series and its order expansion have been tuned for better test accuracy. The performance is validated and compared with previous works for the MNIST dataset. Besides, the digital design of AF is synthesized at 45 nm technology node and physical parameters are compared with previous works. The proposed hardware reused architecture is verified for neural network 16:16:10:4 using 8-bit dynamic fixed-point arithmetic and implemented on Xilinx Zynq xc7z010clg400 SoC using 100 MHz clock. The implemented architecture uses 25% less hardware resources and consumes 12% less power without performance loss, compared to other state-of-the-art implementations, as lower hardware resources and power consumption are especially important for increasingly important edge computing solutions.

A Hybrid Model of Extreme Learning Machine Based on Bat and Cuckoo Search Algorithm for Regression and Multiclass Classification

Extreme learning machine (ELM), as a new simple feedforward neural network learning algorithm, has been extensively used in practical applications because of its good generalization performance and fast learning speed. However, the standard ELM requires more hidden nodes in the application due to the random assignment of hidden layer parameters, which in turn has disadvantages such as poorly hidden layer sparsity, low adjustment ability, and complex network structure. In this paper, we propose a hybrid ELM algorithm based on the bat and cuckoo search algorithm to optimize the input weight and threshold of the ELM algorithm. We test the numerical experimental performance of function approximation and classification problems under a few benchmark datasets; simulation results show that the proposed algorithm can obtain significantly better prediction accuracy compared to similar algorithms.

A new QoS prediction model using hybrid IOWA-ANFIS with fuzzy C-means, subtractive clustering and grid partitioning

Quality of Service (QoS) is one of the key indicators to measure the overall performance of cloud services. The quantitative measurement of the QoS enables the service provider to manage its Service Level Agreement (SLA) in a viable way. It also supports a consumer in service selection and allows measuring the received services to comply with agreed services. There is much existing literature that tries to predict the QoS and assist stakeholders in their decision-making process. However, it is tricky to deal with multidimensional data in time series prediction methods. The computational complexity increases with an increase in data dimension, and it is a challenging task to give precise weights to each time interval. Existing prediction methods could not deal with the intricate reordering of input weights. To address this problem, we propose a novel Clustered Induced Ordered Weighted Averaging (IOWA) Adaptive Neuro-Fuzzy Inference System (ANFIS), (CI-ANFIS) model. This fuzzy time series prediction model reduces data dimension and handles the nonlinear relationship of the cloud QoS dataset. The proposed method uses an intelligent sorting mechanism that regulates uncertainty in prediction while incorporating a fuzzy neural network structure for optimal prediction results. The proposed method employs the IOWA operator to sort input arguments based on associated order-inducing variables and assign customised weights accordingly. The inputs are further classified using three fuzzy clustering methods - fuzzy c-means (FCM), subtractive clustering and grid partitioning. The inputs further pass to the ANFIS structure that takes the benefits of both the fuzzy and neural networks. The fuzzy structure in ANFIS builds understandable rules for cloud stakeholders and deals with uncertain occurrences of data. The model uses a real cloud QoS dataset extracted from the Amazon Elastic Compute Cloud (EC2) US-West instance and predict its behaviour every five minutes for the next 24 h. The proposed method is further compared with the existing twelve methods. The comparative results show that the proposed CI-ANFIS model outperforms all current techniques. The proposed approach opens a new area of research in various complex prediction problems such as stock trading, big data, complex IoT sensors, and other social computing problems.

The Effects of Long Duration Spaceflight on Sensorimotor Control and Cognition.

Astronauts returning from spaceflight typically show transient declines in mobility and balance. Other sensorimotor behaviors and cognitive function have not been investigated as much. Here, we tested whether spaceflight affects performance on various sensorimotor and cognitive tasks during and after missions to the International Space Station (ISS). We obtained mobility (Functional Mobility Test), balance (Sensory Organization Test-5), bimanual coordination (bimanual Purdue Pegboard), cognitive-motor dual-tasking and various other cognitive measures (Digit Symbol Substitution Test, Cube Rotation, Card Rotation, Rod and Frame Test) before, during and after 15 astronauts completed 6 month missions aboard the ISS. We used linear mixed effect models to analyze performance changes due to entering the microgravity environment, behavioral adaptations aboard the ISS and subsequent recovery from microgravity. We observed declines in mobility and balance from pre- to post-flight, suggesting disruption and/or down weighting of vestibular inputs; these behaviors recovered to baseline levels within 30 days post-flight. We also identified bimanual coordination declines from pre- to post-flight and recovery to baseline levels within 30 days post-flight. There were no changes in dual-task performance during or following spaceflight. Cube rotation response time significantly improved from pre- to post-flight, suggestive of practice effects. There was also a trend for better in-flight cube rotation performance on the ISS when crewmembers had their feet in foot loops on the “floor” throughout the task. This suggests that tactile inputs to the foot sole aided orientation. Overall, these results suggest that sensory reweighting due to the microgravity environment of spaceflight affected sensorimotor performance, while cognitive performance was maintained. A shift from exocentric (gravity) spatial references on Earth toward an egocentric spatial reference may also occur aboard the ISS. Upon return to Earth, microgravity adaptions become maladaptive for certain postural tasks, resulting in transient sensorimotor performance declines that recover within 30 days.

Adaptive Cuckoo Search-Extreme Learning Machine Based Prognosis for Electric Scooter System under Intermittent Fault

In this paper, an adaptive Cuckoo search extreme learning machine (ACS-ELM)-based prognosis method is developed for an electric scooter system with intermittent faults. Firstly, bond-graph-based fault detection and isolation is carried out to find possible faulty components in the electric scooter system. Secondly, submodels are decomposed from the global model using structural model decomposition, followed by adaptive Cuckoo search (ACS)-based distributed fault estimation with less computational burden. Then, as the intermittent fault gradually deteriorates in magnitude, and possesses the characteristics of discontinuity and stochasticity, a set of fault features that can describe the intermittent fault’s evolutionary trend are captured with the aid of tumbling window. With the obtained dataset, which represents the fault features, the ACS-ELM is developed to model the intermittent fault degradation trend and predict the remaining useful life of the intermittently faulty component when the physical degradation model is unavailable. In the ACS-ELM, the ACS is employed to optimize the input weights and hidden layer biases of an extreme learning machine, to improve the algorithm performance. Finally, the proposed methodologies are validated by a series of simulation and experiment results based on the electric scooter system.

Lithium-Ion Battery State of Health Estimation Based on Improved Deep Extreme Learning Machine

Abstract Accurate estimation of the state of health (SOH) is an important guarantee for safe and reliable battery operation. In this paper, an online method based on indirect health features (IHFs) and sparrow search algorithm fused with deep extreme learning machine (SSA-DELM) of lithium-ion batteries is proposed to estimate SOH. First, the temperature and voltage curves in the battery discharge data are acquired, and the optimal intervals are obtained by ergodic method. Discharge temperature difference at equal time intervals (DTD-ETI) and discharge time interval with equal voltage difference (DTI-EVD) are extracted as IHF. Then, the input weights and hidden layer thresholds of the DELM algorithm are optimized using SSA, and the SSA-DELM model is applied to the estimation of battery's SOH. Finally, the established model is experimentally validated using the battery data, and the results show that the method has high prediction accuracy, strong algorithmic stability, and good adaptability.

Fabric defect detection based on feature fusion of a convolutional neural network and optimized extreme learning machine

Aiming to accurately detect various defects in the fabric production process, we propose a fabric defect detection algorithm based on the feature fusion of a convolutional neural network (CNN) and optimized extreme learning machine (ELM). Firstly, we use transfer learning to transfer the parameters of the first 13 convolutional layers and first two fully connected layers of a VGG16 network model as pre-trained by ImageNet to the initial model and fine-tune the parameters. Subsequently, the fine-tuned model is used as a feature extractor to extract features of RGB images and their corresponding L-component images. A principal component analysis is used to reduce the dimensionality of the features and fuse the reduced features. The moth flame optimization (MFO) algorithm is used to initialize the optimization variables of a parallel chaotic search (PCS) algorithm, and the PCS algorithm (as optimized by the MFO algorithm) is used to optimize the input weight and bias of the ELM (i.e., the PCS-MFO-ELM (PMELM)). Finally, the PMELM is used to replace the softmax classifier of the CNN to classify and detect fabric defect features. The experimental results show that on the amplified TILDA dataset, the precision, recall, F1-score, and accuracy rates of this algorithm for fabric holes, stains, warp breaks, dragging, and folds in fabric can reach 98.57%, 98.52%, 98.52%, and 98.50%, respectively, that is, higher than those of other algorithms. Through a validity experiment, this method is shown to be suitable for defect detection for unpatterned fabrics, regular patterned fabrics, and irregularly patterned fabrics.

Target Threat Assessment in Air Combat Based on Improved Glowworm Swarm Optimization and ELM Neural Network

Target threat assessment technology is one of the key technologies of intelligent tactical aid decision-making system. Aiming at the problem that traditional beyond-visual-range air combat threat assessment algorithms are susceptible to complex factors, there are correlations between assessment indicators, and accurate and objective assessment results cannot be obtained. A target threat assessment algorithm based on linear discriminant analysis (LDA) and improved glowworm swarm optimization (IGSO) algorithm to optimize extreme learning machine (ELM) is proposed in this paper. Firstly, the linear discriminant analysis method is used to classify the threat assessment indicators, eliminate the correlation between the assessment indicators, and achieve dimensionality reduction of the assessment indicators. Secondly, a prediction model with multiple parallel extreme learning machines as the core is constructed, and the input weights and thresholds of extreme learning machines are optimized by the improved glowworm swarm optimization algorithm, and the weighted integration is carried out according to the training level of the kernel. Then, the threat assessment index functions of angle, speed, distance, altitude, and air combat capability are constructed, respectively, and the sample data of air combat target threat assessment are obtained by combining the structure entropy weight method. Finally, the air combat data is selected from the air combat maneuvering instrument (ACMI), and the accuracy and real-time performance of the LDA-IGSO-ELM algorithm are verified through simulation. The results show that the algorithm can quickly and accurately assess target threats.

Impaired selection of a previously ignored singleton: Evidence for salience map plastic changes.

Spatial suppression of a salient colour distractor is achievable via statistical learning. Distractor suppression attenuates unwanted capture, but at the same time target selection at the most likely distractor location is impaired. This result corroborates the idea that the distractor salience is attenuated via inhibitory signals applied to the corresponding location in the priority map. What is less clear, however, is whether lingering impairment in target selection when the distractor is removed are due to the proactive strategic maintenance of the suppressive signal at the previous most likely distractor location or result from the fact that suppression has induced plastic changes in the priority map, probably changing input weights. Here, we provide evidence that supports the latter possibility, as we found that impairment in target selection persisted even when the singleton distractor in the training phase became the target of search in a subsequent test phase. This manipulation rules out the possibility that the observed impairments at the previous most likely distractor location were caused by a signal suppression maintained at this location. Rather, the results reveal that the inhibitory signals cause long-lasting changes in the priority map, which affect future computation of the target salience at the same location, and therefore the efficiency of attentional selection.

Design of an Adaptive Nonnegative Garrote Algorithm for Multi-Layer Perceptron- Based Soft Sensor

In the study, we propose an adaptive variable selection algorithm for multi-layer perceptron (MLP)-based soft sensors. The proposed algorithm employs nonnegative garrote (NNG) to shrink the input weights of the trained MLP. To improve the shrinkage efficiency of the NNG, adaptive operators are designed using the mean impact value estimate. Moreover, the adaptive operators are data dependent, and are introduced into the constraints of NNG to make the shrinkage more efficient and effective. Cross-validation and Bayesian information criterion are used to determine the optimal garrote parameter for the NNG. The performance of the algorithm is validated using artificial datasets and a practical industrial application in coke dry quenching (CDQ) systems. The simulation results demonstrate that the adaptive mechanism improves the efficiency and precision of NNG, and has superior performance to other state-of-the-art algorithms. The result of variable selection is consistent with the experience of the field operator, and can provide technical supports for the optimisation and control of the CDQ process.

MPC closed-loop identification without excitation

This paper presents a method of closed-loop identification for multivariable systems without external excitation. The method is specially designed for model predictive control (MPC) systems. Without using external excitation (test signals), the method ensures the informativity of the closed-loop data and, at the same time, improve the control performance during the test period. The purpose of the study is to reduce the cost of identification test. The basic idea is to switch the input weighting matrix in the MPC controller which leads to the informativity of the data-set. A preliminary test is carried out in order to find a new input weighting matrix which improve the control performance; then a switching scheme is developed based on the two weighting matrices. Traditional simulation based model validation no longer works in closed-loop identification without excitation, and model error bounds on the frequency responses can be used instead. The effectiveness of the proposed method is demonstrated by a simulation study.

Ultra Short-Term Wind Power Forecasting Based on Sparrow Search Algorithm Optimization Deep Extreme Learning Machine

Improving the accuracy of wind power forecasting is an important measure to deal with the uncertainty and volatility of wind power. Wind speed and wind direction are the most important factors affecting the power generation of wind turbines. In this paper, we propose a wind power forecasting method that combines the sparrow search algorithm (SSA) with the deep extreme learning machine (DELM). Based on the DELM model, the length of the time series’ influence on the performance of the neural network is validated through the comparison of the forecast error indexes, and the optimal time series length of the wind power is determined. The sparrow search algorithm is used to optimize its parameters to solve the problem of random changes in model input weights and thresholds. The proposed SSA-DELM model is validated using the measured data of a certain wind turbine, and various forecasting indexes are compared with several current wind power forecasting methods. The experimental results show that the proposed model has better performance in ultra-short-term wind power forecasting, and its coefficient of determination (R²), mean absolute error (MAE), and root mean square error (RMSE) are 0.927, 69.803, and 115.446, respectively.

Own-race faces promote integrated audiovisual speech information

The other-race effect indicates a perceptual advantage when processing own-race faces. This effect has been demonstrated in individuals' recognition of facial identity and emotional expressions. However, it remains unclear whether the other-race effect also exists in multisensory domains. We conducted two experiments to provide evidence for the other-race effect in facial speech recognition, using the McGurk effect. Experiment 1 tested this issue among East Asian adults, examining the magnitude of the McGurk effect during stimuli using speakers from two different races (own-race vs. other-race). We found that own-race faces induced a stronger McGurk effect than other-race faces. Experiment 2 indicated that the other-race effect was not simply due to different levels of attention being paid to the mouths of own- and other-race speakers. Our findings demonstrated that own-race faces enhance the weight of visual input during audiovisual speech perception, and they provide evidence of the own-race effect in the audiovisual interaction for speech perception in adults.

Broad learning extreme learning machine for forecasting and eliminating tremors in teleoperation

Unwanted errors caused by hand tremors are a bottleneck for the application of teleoperation robots in space explorations, underwater explorations, and minimally invasive surgery. In order to eliminate hand tremor signals in teleoperation control systems, two tremor-filtering models based on artificial neural networks are defined. With the purpose of decreasing the errors of tremor filtering models, a novel Broad Learning Extreme Learning Machine with Improved Equilibrium Optimizer (IEO-BLELM) is proposed. Firstly, the structure of Extreme Learning Machine (ELM) is re-designed by coupling with broad learning. Time series and smoothing are introduced as feature extraction layer and enhancement layer, respectively. Secondly, different activation functions are selected to construct Broad Learning ELM (BLELM). An Improved Equilibrium Optimizer is introduced to optimize input weights, thresholds, and parameters of the BLELM model. To verify the performance of the IEO-BLELM model, the proposed model is applied to six examples and compared with other models. The results show that Mean Absolute Error (MAE) of the proposed model in six examples is at least lower than 0.253. As compared with the ELM, the MAE of the IEO-BLELM model can be decreased by 51.03% through reasonable improvement strategies. In particular, estimation errors are mainly contributed to peak and the proposed model significantly reduces the peak errors. The forecasting performance of the proposed model is better than that of previously existing models. In general, this study provides effective models to eliminate hand tremor signals in teleoperation control systems.

Bayesian stochastic configuration networks for robust data modeling

AbstractThe SCN networks is incrementally generated by stochastic configuration (SC) algorithms. It randomly assigns the input weights and deviations of hidden nodes through a supervisory mechanism, which can be trained by solving linear modeling problems. The version that uses least squares to estimate the output weight performs well. This article introduces an alternative strategy for performing complete Bayesian inference (BI) of SCN networks. Different from the traditional way, the Bayesian training algorithm we proposed can obtain an entire probability distribution on the optimal output weight of the SCN networks, instead of a single pointwise estimate. The advantage of the Bayesian inference method lies in the possibility of introducing other prior knowledge during the training process, providing a way to measure uncertainty during the testing phase and the ability to automatically infer hyper‐parameters from given data. The BI algorithm proposed in this article for regression problem can be implemented through an iterative process under some practical assumptions. Experimental results show that our proposed Bayesian SCN algorithm performs well in solving data modeling problems with a large number of outliers.

Projectile drag coefficient identification based on extreme learning

Aerodynamic parameters play a decisive role in the ballistic characteristics of the projectile. How to accurately obtain the aerodynamic parameters of the projectile is an important task in the development process of the projectile. In order to further improve the identification accuracy of the projectile drag coefficient, this paper generates huge ballistic data through numerical simulation and uses the extreme learning method to identify the ballistic drag coefficient under three kinds of noise conditions. The method avoids the iterative updating process of weights and thresholds by randomly generating the input weights and threshold values of hidden layer neurons and overcomes the problem of long identification time of the traditional back propagation (BP) neural network algorithm. Based on the least squares principle, the Moore–Penrose generalized inverse matrix of the hidden layer output matrix was solved to determine the optimal output weight of the network, and then, the projectile drag coefficient was accurately identified. Comparing the extreme learning method with the traditional BP neural network method, the results show that the proposed method has higher identification accuracy and faster convergence speed and can effectively identify the projectile drag coefficient, which can meet the practical needs of engineering.

High-yield hydrogen from thermal processing of waste plastics

The production of hydrogen (H2) from the pyrolysis–catalytic steam reforming of polyethylene, polystyrene (PS) and polyethylene terephthalate waste plastics was investigated using a two-stage reactor. The highest yield of hydrogen (125 mmol/gplastic) was obtained with PS at a catalyst temperature of 900°C and steam input weight hourly space velocity of 7.59 g/(h/gcatalyst) with a 10 wt% nickel/aluminium oxide (Ni/Al2O3) catalyst. Further investigation using PS showed that the process parameters of high catalyst temperature (900°C) and optimised steam input rate significantly increased the yield of hydrogen. Examination of several different catalysts (nickel/aluminium oxide, iron/aluminium oxide, copper/aluminium oxide, cobalt/aluminium oxide) showed that nickel/aluminium oxide had by far the highest catalytic activity and selectivity towards the yield of hydrogen.

Organizational principles of amygdalar input-output neuronal circuits

The amygdala, one of the most studied brain structures, integrates brain-wide heterogeneous inputs and governs multidimensional outputs to control diverse behaviors central to survival, yet how amygdalar input-output neuronal circuits are organized remains unclear. Using a simplified cell-type- and projection-specific retrograde transsynaptic tracing technique, we scrutinized brain-wide afferent inputs of four major output neuronal groups in the amygdalar basolateral complex (BLA) that project to the bed nucleus of the stria terminals (BNST), ventral hippocampus (vHPC), medial prefrontal cortex (mPFC) and nucleus accumbens (NAc), respectively. Brain-wide input-output quantitative analysis unveils that BLA efferent neurons receive a diverse array of afferents with varied input weights and predominant contextual representation. Notably, the afferents received by BNST-, vHPC-, mPFC- and NAc-projecting BLA neurons exhibit virtually identical origins and input weights. These results indicate that the organization of amygdalar BLA input-output neuronal circuits follows the input-dependent and output-independent principles, ideal for integrating brain-wide diverse afferent stimuli to control parallel efferent actions. The data provide the objective basis for improving the virtual reality exposure therapy for anxiety disorders and validate the simplified cell-type- and projection-specific retrograde transsynaptic tracing method.