Functional Link Neural Network Research Articles

Recurrent ensemble random vector functional link neural network for financial time series forecasting

Financial time series forecasting is crucial in empowering investors to make well-informed decisions, manage risks effectively, and strategically plan their investment activities. However, the non-stationary and non-linear characteristics inherent in time series data pose significant challenges when accurately predicting future forecasts. This paper proposes a novel Recurrent ensemble deep Random Vector Functional Link (RedRVFL) network for financial time series forecasting. The proposed model leverages randomly initialized and fixed weights for the recurrent hidden layers, ensuring stability during training. Furthermore, incorporating stacked hidden layers enables deep representation learning, facilitating the extraction of complex patterns from the data. The proposed model generates the forecast by combining the outputs of each layer through an ensemble approach. A comparative analysis was conducted against several state-of-the-art models over financial time-series datasets, and the results demonstrated the superior performance of our proposed model in terms of forecasting accuracy and predictive capability.

Neuro-Fuzzy Random Vector Functional Link Neural Network for Classification and Regression Problems

Neuro-Fuzzy Random Vector Functional Link Neural Network for Classification and Regression Problems

Graph Embedded Intuitionistic Fuzzy Random Vector Functional Link Neural Network for Class Imbalance Learning.

The domain of machine learning is confronted with a crucial research area known as class imbalance (CI) learning, which presents considerable hurdles in the precise classification of minority classes. This issue can result in biased models where the majority class takes precedence in the training process, leading to the underrepresentation of the minority class. The random vector functional link (RVFL) network is a widely used and effective learning model for classification due to its good generalization performance and efficiency. However, it suffers when dealing with imbalanced datasets. To overcome this limitation, we propose a novel graph-embedded intuitionistic fuzzy RVFL for CI learning (GE-IFRVFL-CIL) model incorporating a weighting mechanism to handle imbalanced datasets. The proposed GE-IFRVFL-CIL model offers a plethora of benefits: 1) leveraging graph embedding (GE) to preserve the inherent topological structure of the datasets; 2) employing intuitionistic fuzzy (IF) theory to handle uncertainty and imprecision in the data; and 3) the most important, it tackles CI learning. The amalgamation of a weighting scheme, GE, and IF sets leads to the superior performance of the proposed models on KEEL benchmark imbalanced datasets with and without Gaussian noise. Furthermore, we implemented the proposed GE-IFRVFL-CIL on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset and achieved promising results, demonstrating the model's effectiveness in real-world applications. The proposed GE-IFRVFL-CIL model offers a promising solution to address the CI issue, mitigates the detrimental effect of noise and outliers, and preserves the inherent geometrical structures of the dataset.

Multiobjective Operation Optimization of Wastewater Treatment Process Based on Reinforcement Self-Learning and Knowledge Guidance.

This article proposes a multiobjective operation optimization method based on reinforcement self-learning and knowledge guidance for quality assurance and consumption reduction of wastewater treatment process (WWTP) with nonstationary time-varying dynamics. First, operation optimization models are developed by online sequential random vector functional-link (OS-RVFL) neural network, which can realize online sequential learning of model parameters. Then, a knowledge base is established to store typical optimization cases for knowledge guiding the subsequent optimizations. Based on it, a reinforcement self-learning-based multiobjective particle swarm optimization (RSL-MOPSO) algorithm is proposed to perform optimization calculation. In this algorithm, reinforcement self-learning is used for interaction learning between environment and action in optimization, and the particle motion trend of algorithm is adjusted according to the feedback information of the optimization process. The effects of wastewater state parameters on particles are recorded and reused to improve the solution quality and calculation efficiency of optimization. Moreover, to make good use of the information of the previous optimizations and balance the coordination between global search in the early stage and local search in the later stage, a selective information feedback mechanism is further proposed to ensure the diversity and convergence of the algorithm. Finally, prediction-based intelligent decision making is performed to select the final optimization solution as the final setpoints for the lower-level controllers from the Pareto frontier with considering specific technical requirements. Data experiments show that the proposed method can effectively reduce energy consumption and ensure effluent quality.

Denoising or distortion: Does decomposition-reconstruction modeling paradigm provide a reliable prediction for crude oil price time series?

Reliable forecasting of the crude oil price has important implications for production planning, investment decisions and policy making. Though the effectiveness of denoising hybrid models in crude oil price forecasting has been widely demonstrated, we argue that the denoised series used for model fitting may be affected by the out-of-sample information due to the one-time denoising operation applied to the entire price series. To verify whether the denoising paradigm does improve the crude oil price forecasting, a rolling window mechanism is introduced into two denoising hybrid models based on ensemble empirical mode decomposition (EEMD) and variational mode decomposition (VMD). The novel R-EEMD-RVFL and R-VMD-RVFL models with the random vector functional link (RVFL) neural network are proposed. Then, both monthly and weekly West Texas Intermediate crude oil spot prices are used for multi-step-forward prediction. The experiment reports that: Firstly, the prediction provided by conventional denoising hybrid models is unreliable due to the one-time denoising procedure. Secondly, the proposed R-VMD-RVFL model achieves superior and robust accuracies in both monthly and weekly crude oil price predictions, especially when the forecast horizon is relatively long. Furthermore, this study provides theoretical supports for the findings using the weak-form efficient market hypothesis and behavioral finance theory as well as technical analysis from the perspective of denoising methods.

An improved proportionate arctangent framework based LMS/F algorithm for system identification

This paper proposes an improved proportionate arctangent framework that relies on a least mean square/fourth (IPALMS/F) algorithm for the physical sparse system identification problem. The ALMS/F algorithm has significant robustness against impulsive noise, whereas the improved proportionate concept when utilized with the ALMS/F takes advantage of the sparse characteristic to increase convergence time. Finally, the IPALMS/F algorithm is implemented using a recursive adaptive sparse exponential functional link neural network (RASETFLN) nonlinear filter and is named the RASETFLN-IPALMS/F algorithm which resulted in enhanced performance compared to other existing filters in terms of robustness in an impulsive environment, convergence rate, and steady-state error for system identification and acoustic echo cancellation.

A Virtual Sensing Active Noise Control System Based on a Functional Link Neural Network for an Aircraft Seat Headrest

PurposeThe goal of this work is to propose a local Active Noise Control system for an aircraft seat’s headrest. This system should be able to extend the quiet zone beyond the physical microphones that can be placed behind the passenger’s head due to space constraints. Furthermore, the control algorithm should perform efficiently when non-linear phenomena exist during the sound propagation.MethodsA functional link neural network and a computationally efficient Multiple Input Multiple Output approach are used for the system’s implementation. The quiet zone is extended by linearly estimating the acoustic pressure in front of the headrest surface. The main novelty of this paper is the combination of these methods in an attempt to improve acoustic pressure attenuation performance, while keeping computational complexity low.ResultsThe proposed control algorithm has been evaluated through numerical simulations, including Finite Element Method and experimental tests at an aircraft cabin mock-up. The results show that for a real-world acoustic disturbance, a 10 dB reduction in sound pressure level was achieved 10 cm away from the headrest surface. In addition, the attenuation of some harmonics can reach 20 dB and in most cases is bigger than the linear FxLMS algorithm.ConclusionsTo summarize, it has been demonstrated that a multichannel functional link neural network using a simple virtual sensing technique can efficiently attenuate synthesized and real world acoustic disturbances captured in a tilt-rotor aircraft’s cabin. Finally, it can create an adequate quiet zone for gentle head movement while maintaining its stability.

Online dynamic ensemble deep random vector functional link neural network for forecasting

This paper proposes a three-stage online deep learning model for time series based on the ensemble deep random vector functional link (edRVFL). The edRVFL stacks multiple randomized layers to enhance the single-layer RVFL’s representation ability. Each hidden layer’s representation is utilized for training an output layer, and the ensemble of all output layers forms the edRVFL’s output. However, the original edRVFL is not designed for online learning, and the randomized nature of the features is harmful to extracting meaningful temporal features. In order to address the limitations and extend the edRVFL to an online learning mode, this paper proposes a dynamic edRVFL consisting of three online components, the online decomposition, the online training, and the online dynamic ensemble. First, an online decomposition is utilized as a feature engineering block for the edRVFL. Then, an online learning algorithm is designed to learn the edRVFL. Finally, an online dynamic ensemble method, which can measure the change in the distribution, is proposed for aggregating all layers’ outputs. This paper evaluates and compares the proposed model with state-of-the-art methods on sixteen time series.

A novel lagrange functional link neural network for solving variable-order fractional time-varying delay differential equations: a comparison with multilayer perceptron neural networks

A novel lagrange functional link neural network for solving variable-order fractional time-varying delay differential equations: a comparison with multilayer perceptron neural networks

Improving the classification accuracy of fishes and invertebrates using residual convolutional neural networks

Abstract The visibility of fishes and invertebrates is highly impacted by the complexity of the environment. Images acquired in underwater environments suffer from blurriness and low contrast. This results in a low classification accuracy. To address this problem, this study uses a pre-trained Resnet50 neural network as the feature extractor, which avoids over-fitting and accuracy saturation while realizing improved feature extraction capabilities. It also proposes an enhancement of the error-minimized random vector functional link (EEMRVFL) neural network, which is used as the classifier in the convolutional neural network (CNN) model instead of the original softmax classifier. EEMRVFL reduces the maximum residual error in each incremental process. The selected hidden nodes are added to the network, which improves the compactness of its structure. The proposed residual CNNs model exhibits improved classification accuracy for underwater image classification compared to existing methods. This is demonstrated experimentally on available datasets such as URPC, LifeCLEF 2015, and Fish4Knowledge with accuracy rates reaching 99.68%, 97.34%, and 99.77%, respectively.

Limited information limits accuracy: Whether ensemble empirical mode decomposition improves crude oil spot price prediction?

A reliable crude oil price forecast is important for market pricing. Despite the widespread use of ensemble empirical mode decomposition (EEMD) in financial time series forecasting, the one-time decomposition on the entire time series leads the in-sample data to be affected by the out-of-sample data. Consequently, the forecasting accuracy is overstated. This study incorporates a rolling window into two prevalent EEMD-based modeling paradigms, namely decomposition-ensemble and denoising, to ensure that only in-sample time series is processed by EEMD and used for model training. Given the time-consuming process of stepwise preprocessing and model fitting, two non-iterative machine learning algorithms, random vector functional link (RVFL) neural network and extreme learning machine (ELM), are used as predictors. Hence, we develop the rolling decomposition-ensemble and rolling denoising paradigms, respectively. Contrary to the majority of prior studies, empirical results based on monthly spot price time series for the Brent and West Texas Intermediate (WTI) markets indicate that EEMD plays a weak role in improving crude oil price forecasts when only the in-sample set is preprocessed. This is compatible with the weak form of the efficient market hypothesis (EMH). Nevertheless, the suggested rolling EEMD-denoising model has an advantage over other employed models for long-term forecasting.

An Intelligent Telugu Handwritten Character Recognition Using Multi-Objective Mayfly Optimization with Deep Learning–Based DenseNet Model

The handwritten character recognition process has gained significant attention among research communities due to its application in assistive technologies for visually impaired people, human–robot interaction, automated registry for business documents, and so on. Handwritten character recognition of the Telugu language is difficult owing to the absence of a massive dataset and a trained convolutional neural network (CNN). This article introduces an intelligent Telugu character recognition process using a multi-objective mayfly optimization with deep learning (MOMFO-DL) model. The proposed MOMFO-DL technique involves the DenseNet-169 model as a feature extractor to generate a useful set of feature vectors. A functional link neural network (FLNN) is used as a classification model to recognize and classify the printer characters. The design of the MOMFO technique for the parameter optimization of the DenseNet model and FLNN model shows the novelty of the work. The use of MOMFO technique helps to optimally tune the parameters in such a way that the overall performance can be improved. The extensive experimental analysis takes place on benchmark datasets and the outcomes are examined with respect to different measures. The experimental results pointed out the supremacy of the MOMFO technique over the recent state-of-the-art methods.

A Bidirectional Onboard Charger With Multistep Constant Current Charging Capability

This article realizes a bidirectional onboard charger (OBC) for light electric vehicles (LEVs), with multistep-constant-current (MSCC) charging capability. In this MSCC charging operation, the battery current is reduced during the peak-load hours and resumes the rated charging operation amid the off-peak load hours. Since this MSCC operation is a wide load range operation, the controllers of the front-end converter (FEC) and back-end converter are modified. In this context, a Hermite polynomial-derived functional link neural network (H-FLNN)-based deadbeat predictive current controller (DPCC) is introduced to improve the performance of the FEC of the OBC. Moreover, the HFLNN-based DPCC for the FEC regulates the grid current quality during MSCC and distinct operating conditions such as grid voltage sag and swell, distorted grid voltage, and the presence of dc offset error in the measured grid voltage. A laboratory prototype of a 1.1-kW rating is developed to verify the efficacy of the H-FLNN DPCC to keep the grid current power quality (PQ) within the International Electrotechnical Commission (IEC)-61000-3-2 standard throughout the operation. Moreover, the effectiveness of the design is verified amid the grid voltage sag and swell and distinct operating conditions. The experimental analysis depicts that the rated efficiency of the developed laboratory prototype is 90.18% and the analytical power density is 64.16 W/in3. Furthermore, the vehicle to grid (V2G) operation validates the satisfactory design of OBC for bidirectional operation. At last, the comparison of the presented system with existing systems is given to summarize the advantages of the presented system over the existing systems.

Factorization of broad expansion for broad learning system

The broad learning system (BLS) based on the random vector functional link neural network is a new versatile non-iterative neural network for rapidly selecting models. One of its fascinating features is the broad expansion, a dynamic adjustment strategy in the changing environment. However, the added enhancement nodes often suffer from information redundancy and loss, a failure in the broad expansion. Therefore, the factorization of enhancement nodes, which makes each code independent of others and achieves minimal redundancy and loss representation, is critical to successful broad expansion. Inspired by this idea, in this study, a factorial broad expansion (FBE) strategy is proposed to aid BLS (FBE-BLS) during broad expansion. The strategy is proposed as a component for reducing the information redundancy and loss in the classical BLS. This strategy is an adversarial framework to learn the factorial codes at the enhancement layer. The factorial and predictor networks adversarially learn factorial codes by chasing independence, minimizing predictability among enhancement nodes. In experiments, considering that few works focused on information redundancy and loss in the broad expansion of BLS, the FBE-BLS is compared with the classical BLS, and state-of-the-art methods are selected. The results of FBE-BLS on UCI and real-world image data are very promising, exhibiting higher accuracy during broad expansion.

Electronic dispersion compensation and functional link neural network equalization for IM/DD links

Short-reach fiber optical links employing intensity modulation (IM) at the transmitter (Tx) and direct detection (DD) at the receiver (Rx), suffer from linear and nonlinear sources of impairments due to the interaction of chromatic dispersion (CD) with DD. In this article, joint electronic dispersion compensation (EDC) at the Tx, using two distinct Gerchberg–Saxton(GS) based approaches, and at the Rx, using a functional link neural network (FLNN) equalizer is demonstrated for IM/DD transmission. The first Tx approach utilizes the modified iterative GS algorithm, which partially mitigate linear and nonlinear sources of inter-symbol interference (ISI). The second Tx pre-EDC approach only pre-compensates for the linear power fading effect through implementing a GS based finite impulse response (FIR) filter. At the Rx, a T/2-spaced or T-spaced adaptive post-feed forward equalizer (FFE) is employed to fully compensate residual chromatic dispersion prior to attempting nonlinear equalization. Furthermore, a Volterra nonlinear equalizer (VNLE) is introduced to benchmark the performance and complexity of the FLNN, Subsequently, either a FLNN or a VNLE are utilized for nonlinear system identification and subsequent post-equalization mitigating uncompensated nonlinear sources of ISI. The FLNN nonlinear taps resulting from the functional expansion block are optimized using the recursive least square (RLS) algorithm. The Tx-FIR and Rx-FLNN enable 112 Gbit/s non-return to zero (NRZ) on–off keying (OOK) transmission over 20 km of single mode fiber (SMF) and 112 Gbit/s 4-level pulse-amplitude modulation (PAM-4) transmission over 10 km of SMF. It is shown that the use of Tx pre-EDC reduces the complexity of the required equalization at the Rx. In addition, the FLNN was found to offer a 74% reduction in computational complexity relative to the third-order VNLE.

Water distillation tower: Experimental investigation, economic assessment, and performance prediction using optimized machine-learning model

Here we present a new compact design of a vertical water distillation tower based on solar stills. The experimental setup consisted of a vertical tower with five water trays, supported by a metal duct and surrounded by a glass enclosure. The water yield and thermal, exergic, and economic features of the tower were investigated and analyzed. A new performance prediction hybrid model was also developed. A powerful artificial intelligence tool called the random vector functional link (RVFL) neural network was integrated with the Runge Kutta optimizer (RUN) to predict the water yield and temperature of the established tower. The model efficiency was compared to that of pure RVFL and an optimized RVFL model using a particle swarm optimizer (PSO). The drinkable water yield of the proposed design was 2.1 L/m2 (considering the tray area) and 5.3 L/m2 (considering the land use area); energy and exergy efficiencies were ∼31.7% and ∼3.3%, respectively. The cost of the produced drinkable water was approximately $0.013/L. The developed system provides considerable improvement compared with conventional designs of solar stills. The proposed RVFL–RUN model outperformed the pure RVFL and RVFL–PSO models for predicting system performance. The coefficients of determination between the experimental water productivity and water temperature and the predicted values, using the RVFL–RUN model, were 0.91 and 0.97, respectively.

Brain Tumor Diagnosis Using Sparrow Search Algorithm Based Deep Learning Model

Recently, Internet of Medical Things (IoMT) has gained considerable attention to provide improved healthcare services to patients. Since earlier diagnosis of brain tumor (BT) using medical imaging becomes an essential task, automated IoMT and cloud enabled BT diagnosis model can be devised using recent deep learning models. With this motivation, this paper introduces a novel IoMT and cloud enabled BT diagnosis model, named IoMTC-HDBT. The IoMTC-HDBT model comprises the data acquisition process by the use of IoMT devices which captures the magnetic resonance imaging (MRI) brain images and transmit them to the cloud server. Besides, adaptive window filtering (AWF) based image preprocessing is used to remove noise. In addition, the cloud server executes the disease diagnosis model which includes the sparrow search algorithm (SSA) with GoogleNet (SSA-GN) model. The IoMTC-HDBT model applies functional link neural network (FLNN), which has the ability to detect and classify the MRI brain images as normal or abnormal. It finds useful to generate the reports instantly for patients located in remote areas. The validation of the IoMTC-HDBT model takes place against BRATS2015 Challenge dataset and the experimental analysis is carried out interms of sensitivity, accuracy, and specificity. The experimentation outcome pointed out the betterment of the proposed model with the accuracy of 0.984.

Functional Link Neural Network for Wide Load Operation of Bidirectional Onboard Charging System of E-Rickshaw

This article investigates the functional link neural network assisted deadbeat predictive current control for an onboard charger (OBC) of an e-rickshaw with multistep charging capability. This control algorithm is presented for wide load operation of OBC since a multistep charging scheme is introduced to reduce the electricity consumption by OBC in the course of the peak load hours. The presented grid-connected OBC topology has two stages interconnected with the dc link capacitor. The first stage is a front-end converter to regulate the voltage at the dc link capacitor. Moreover, it regulates the power quality within the IEEE-519 standard. Furthermore, a dc–dc converter to regulate the charging current is the second stage of the OBC. Keeping this in view, a bidirectional interleaved isolated single ended primary inductor converter (SEPIC) converter topology is utilized. To analyze the performance of the OBC under distinct operating conditions such as grid voltage variations, an experimental prototype of the bidirectional OBC, is developed. Furthermore, the efficacy of the bidirectional operation of the designed system is analyzed through the vehicle to grid operation.

Design of delayless multi-sampled subband functional link neural network with application to active noise control

To accelerate the convergence speed of the functional link neural network (FLNN) particularly for colored input signals, this paper proposes a delayless multi-sampled multiband-structured subband FLNN (DMSFLNN) structure. Then, to update the weights of the DMSFLNN, a normalized subband adaptive algorithm is devised. Next, the stability conditions, optimal step size and computational complexity are investigated. Moreover, the proposed method is applied to the nonlinear active noise control, obtaining the delayless multi-sampled multiband-structured filtered-s normalized least mean square (DMSFsNLMS) algorithm. Finally, simulation results demonstrate that the proposed method improves the convergence speed of the FLNN.

Do EEMD based decomposition-ensemble models indeed improve prediction for crude oil futures prices?

The ensemble empirical mode decomposition (EEMD) based decomposition-ensemble models are widely applied to crude oil futures prices prediction. However, we argue whether EEMD really improves the prediction as the one-time decomposition would cause the exposure of future features, which may result in a misleadingly high accuracy. Therefore, a sliding decomposition-ensemble paradigm SW-EEMD-RVFL is proposed, which conducts the decomposition only of historical series in the sliding window for each forecasting iteration. The random vector functional link (RVFL) neural network is employed as the forecasting approach. The Brent and West Texas Intermediate (WTI) crude oil futures are taken to verify the model by comprehensively comparing individual econometric, artificial intelligence models and their hybrid forms. The findings show that, although hybrid models perform well in the in-sample data, SW-EEMD based models cannot outperform neither the EEMD based models or individual models in the out-of-sample data. EEMD cannot improve the prediction when only based on the decomposition of historical series, which is not consistent with most researches. Except only focusing on the improving accuracy, this work also gives possible explanations of the forecasting results from both methodologies and theories, as well as demonstrates the Efficient Markets Hypothesis (EMH) in terms of artificial intelligence methods.