Self-organizing Neural Network Research Articles

Emergence and Criticality in Spatiotemporal Synchronization: The Complementarity Model.

This work concerns the long-term collective excitability properties and the statistical analysis of the critical events displayed by a recently introduced spatiotemporal many-body model, proposed as a new paradigm for Artificial Life. Numerical simulations show that excitable collective structures emerge in the form of dynamic networks, created by bursts of spatiotemporal activity (avalanches) at the edge of a synchronization phase transition. The spatiotemporal dynamics is portraited by a movie and quantified by time varying collective parameters, showing that the dynamic networks undergo a "life cycle," made of self-creation, homeostasis, and self-destruction. The power spectra of the collective parameters show 1/f power law tails. The statistical properties of the avalanches, evaluated in terms of size and duration, show power laws with characteristic exponents in agreement with those values experimentally found in the neural networks literature. The mechanism underlying avalanches is argued in terms of local-to-collective excitability. The connections that link the present work to self-organized criticality, neural networks, and Artificial Life are discussed.

Self-Organizing Type-2 Fuzzy Double Loop Recurrent Neural Network for Uncertain Nonlinear System Control.

Nonlinear systems, such as robotic systems, play an increasingly important role in our modern daily life and have become more dominant in many industries; however, robotic control still faces various challenges due to diverse and unstructured work environments. This article proposes a double-loop recurrent neural network (DLRNN) with the support of a Type-2 fuzzy system and a self-organizing mechanism for improved performance in nonlinear dynamic robot control. The proposed network has a double-loop recurrent structure, which enables better dynamic mapping. In addition, the network combines a Type-2 fuzzy system with a double-loop recurrent structure to improve the ability to deal with uncertain environments. To achieve an efficient system response, a self-organizing mechanism is proposed to adaptively adjust the number of layers in a DLRNN. This work integrates the proposed network into a conventional sliding mode control (SMC) system to theoretically and empirically prove its stability. The proposed system is applied to a three-joint robot manipulator, leading to a comparative study that considers several existing control approaches. The experimental results confirm the superiority of the proposed system and its effectiveness and robustness in response to various external system disturbances.

3D clustering of gene expression data from systemic autoinflammatory diseases using self-organizing maps (Clust3D)

Background and objectiveSystemic autoinflammatory diseases (SAIDs) are characterized by widespread inflammation, but for most of them there is a lack of specific biomarkers for accurate diagnosis. Although a number of machine learning algorithms have been used to analyze SAID datasets, aiding in the discovery of novel biomarkers, there is a growing recognition of the importance of SAID timeseries clustering, as it can capture the temporal dynamics of gene expression patterns. MethodologyThis paper proposes a novel clustering methodology to efficiently associate three-dimensional data. The algorithm utilizes competitive learning to create a self-organizing neural network and adjust neuron positions in time-dependent and high dimensional feature space in order to assign them as clustering centers. The quantitative evaluation of the clustering was based on well-known clustering indices. Furthermore, a differential expression analysis and classification pipeline was employed to assess the capability of the proposed methodology to extract more accurate pathway-specific genes from its clusters. For that, a comparative analysis was also conducted against a heuristic timeseries clustering method. ResultsThe proposed methodology achieved better overall clustering indices scores and classification metrics using genes derived from its clusters. Notable cases include a threefold increase in the Calinski-Harabasz clustering index, a twofold improvement in the Davies–Bouldin clustering index and a ∼60% increase in the classification specificity score. ConclusionA novel clustering methodology was developed and applied on several gene expression timeseries datasets from systemic autoinflammatory diseases, and its ability to efficiently produce well separated clusters compared to existing heuristic methods was demonstrated.

Modeling and diagnosis of water quality parameters in wastewater treatment process based on improved particle swarm optimization and self-organizing neural network

The study designs a strategy to predict water quality parameters and identify faults in the activated sludge process. The proposed method combines the improved beetle antennae search (IBAS) with the dynamic particle swarm multi-objective optimization algorithm based on crowding distance (DMOPSO-CD), named DMOPSO-CD-IBAS. The fault diagnosis prediction interval of chemical oxygen demand (COD), biochemical oxygen demand (BOD5) and total suspended substance (TSS) was conducted by self-organizing fuzzy neural network with an efficient scheme for parsimonious (SOFNN-ESP). DMOPSO-CD enhances the efficiency of the beetle antenna search by adjusting inertia weights and acceleration factors, reaching the optimal state with a fitness value of 0.0961. Furthermore, the crowding distance sorting preserve the external elite set and update the global optimal values, upgrading the individual search mechanism of beetle antenna search to a group search mechanism to enhance population diversity. The proposed fault diagnosis method's effectiveness is validated through three specific faults (sensor fault of COD and BOD5, and the index of sludge expansion) during the sewage treatment process. The results of the prediction mean square error, root mean square error, and average absolute error of the proposed model were kept at about 0.011, 0.15 and 0.524, respectively, having better tracking performance and stronger robustness than SOFNN-ESP and backpropagation.

A hybrid model based on unsupervised learning for seismic response feature extraction and gas-bearing reservoir prediction using longitudinal and transverse waves seismic data in low-degree exploration areas: A case study

Predicting gas-bearing reservoirs from seismic data is a major field of research. Recently, deep learning has been applied to achieve better results; however, challenges such as high sample demand remain. Drilling data are often insufficient, and gas-bearing sample data are difficult to acquire in large quantities, particularly for low-level (new) exploration areas, which limits gas-bearing reservoir prediction by supervised learning. Therefore, this study proposes a hybrid model using unsupervised learning to extract seismic response features and predict gas-bearing reservoir distribution. First, seismic attribute analysis was used to extract as many longitudinal and transverse waves seismic attributes as possible to provide richer information about gas-bearing reservoirs. However, abundant seismic attributes contain a large amount of redundant information, thereby increasing computation time. Thus, principal component analysis (PCA) was used to reduce the dimensions of the extracted multi-component seismic attributes, remove redundant information, and obtain the most important characteristics that are sensitive to gas-bearing reservoirs. The sensitive seismic attributes obtained after the PCA were fed into a self-organizing neural network (SOM) for unsupervised learning. Neurons in the SOM compete with each other to optimize their own features and extract reservoir-sensitive seismic response features for prediction. To evaluate the proposed method, it was applied to part of the Marmousi2 model and real seismic data for validation. The following results were obtained: (1)The results of the synthetic data test results not only proved the effectiveness of the proposed method, but also showed that the method has strong noise resistance capability and is suitable for predicting gas-bearing reservoirs with low signal-to-noise ratio seismic data; (2) Based on the drilling data and sedimentary facies characteristics in the test area, the actual 3D seismic data test results showed that there are four wells that match the predicted results, and the distribution boundaries of the gas-bearing reservoirs they depict are relatively clear; (3) In low exploration areas (areas with few or no wells), this method is important as a guide to hydrocarbon exploration.

HiSOMA: A hierarchical multi-agent model integrating self-organizing neural networks with multi-agent deep reinforcement learning

Multi-agent deep reinforcement learning (MADRL) has shown remarkable advancements in the past decade. However, most current MADRL models focus on task-specific short-horizon problems involving a small number of agents, limiting their applicability to long-horizon planning in complex environments. Hierarchical multi-agent models offer a promising solution by organizing agents into different levels, effectively addressing tasks with varying planning horizons. However, these models often face constraints related to the number of agents or levels of hierarchies. This paper introduces HiSOMA, a novel hierarchical multi-agent model designed to handle long-horizon, multi-agent, multi-task decision-making problems. The top-level controller, FALCON, is modeled as a class of self-organizing neural networks (SONN), designed to learn high-level decision rules as internal cognitive codes to modulate middle-level controllers in a fast and incremental manner. The middle-level controllers, MADRL models, in turn receive modulatory signals from the higher level and regulate bottom-level controllers, which learn individual action policies generating primitive actions and interacting directly with the environment. Extensive experiments across different levels of the hierarchical model demonstrate HiSOMA’s efficiency in tackling challenging long-horizon problems, surpassing a number of non-hierarchical MADRL approaches. Moreover, its modular design allows for extension into deeper hierarchies and application to more complex tasks with heterogeneous controllers. Demonstration videos and codes can be found on our project web page: https://smu-ncc.github.io.

Unveiling the evolution of urban rail transit network: considering ridership attributes

ABSTRACT Previous research on urban rail transit (URT) evolution mainly focused on network topology, neglecting ridership attributes. This study extracts ridership and network topology indicators from Chinese URT data. Employing a self-organizing mapping neural network model, it divides China’s URT development into four stages. The initial stage and the development stage form the framework of URT network. The network diameter reaches the maximum in the networked operation stage. In the mature stage, URT network densification occurs alongside a significant increase in resident ridership. It is also found that each network indicator has a significant nonlinear relationship with ridership attributes. These findings are of guiding significance for urban planners to accurately understanding URT’s future development and rational network planning and construction.

Soft Sensor Modeling of Self-Organizing Interval Type-2 Fuzzy Neural Network Based on Adaptive Quantum-Behaved Particle Swarm Optimization Algorithm

Soft Sensor Modeling of Self-Organizing Interval Type-2 Fuzzy Neural Network Based on Adaptive Quantum-Behaved Particle Swarm Optimization Algorithm

Unsupervised Clustering-Assisted Method for Consensual Quantitative Analysis of Methanol-Gasoline Blends by Raman Spectroscopy.

Methanol-gasoline blends have emerged as a promising and environmentally friendly bio-fuel option, garnering widespread attention and promotion globally. The methanol content within these blends significantly influences their quality and combustion performance. This study explores the qualitative and qualitative analysis of methanol-gasoline blends using Raman spectroscopy coupled with machine learning methods. Experimentally, methanol-gasoline blends with varying methanol concentrations were artificially configured, commencing with initial market samples. For qualitative analysis, the partial least squares discriminant analysis (PLS-DA) model was employed to classify the categories of blends, demonstrating high prediction performance with an accuracy of nearly 100% classification. For the quantitative analysis, a consensus model was proposed to accurately predict the methanol content. It integrates member models developed on clustered variables, using the unsupervised clustering method of the self-organizing mapping neural network (SOM) to accomplish the regression prediction. The performance of this consensus model was systemically compared to that of the PLS model and uninformative variable elimination (UVE)-PLS model. Results revealed that the unsupervised consensus model outperformed other models in predicting the methanol content across various types of methanol gasoline blends. The correlation coefficients for prediction sets consistently exceeded 0.98. Consequently, Raman spectroscopy emerges as a suitable choice for both qualitative and quantitative analysis of methanol-gasoline blend quality. This study anticipates an increasing role for Raman spectroscopy in analysis of fuel composition.

Quantum self-organizing feature mapping neural network algorithm based on Grover search algorithm

Self-organizing feature mapping neural network is a typical unsupervised neural network algorithm, which is often used for clustering analysis and data compression. As the amount of data increases, the time consumption required by the algorithm becomes increasingly large, which becomes a new challenge. To address this issue, a quantum self-organizing feature mapping neural network is proposed in this paper. This algorithm provides a method to obtain the similarity between samples and neurons based on quantum phase estimation and demonstrates the scheme to obtain winning neurons by Grover algorithm. By utilizing the superposition of quantum, the algorithm achieves parallel computing. The time complexity analysis indicates that the proposed algorithm is exponentially faster than the classical counterpart. The quantum circuit has been devised, while numerical simulation and experiment on a heart disease dataset have been conducted programming within the Qiskit framework. Both have verified the feasibility of the algorithm. Moreover, an application of classification has been developed based on the trained self-organizing feature mapping neural network, which demonstrates the effectiveness of the proposed algorithm.

Estimating Brazilian Tensile Strength of Granite Rocks Using Metaheuristic Algorithms-Based Self-Organizing Neural Networks

Estimating Brazilian Tensile Strength of Granite Rocks Using Metaheuristic Algorithms-Based Self-Organizing Neural Networks

Organizational changes and research performance: A multidimensional assessment

Abstract This paper analyzes the research performance evolution of a scientific institute, from its genesis through various stages of development. The main aim is to obtain, and visually represent, bibliometric evidence of the correlation of organizational changes on the development of its scientific performance; particularly, structural and leadership changes. The study involves six bibliometric indicators to multidimensionally assess the evolution of the institution’s performance profile. For a case study, we selected the Renewable Energy Institute at the National Autonomous University of Mexico, created 35 years ago as a small laboratory, then it evolved to a research center and finally to a formal institute, which over the last 8 years changed from the traditional departmental structure to a network-based structure. The evolution of the multidimensional performance profiles is analyzed, and graphically represented, using a novel artificial intelligence-based approach. We analyzed the performance profiles evolution yearly, using Principal Components Analysis, and a self-organizing neural network mapping technique. This approach, combining bibliometric and machine learning techniques, proved to be effective for the assessment of the institution’s evolution process. The results were represented with a series of graphs and maps that clearly reveal the magnitude and nature of the performance profile evolution, as well as its correlation with each of the structural and leadership transitions. These exploratory results have provided us data and insights into the probable effects of these transitions on academic performance, that have been useful to create a dynamical model.

Data-Knowledge-Driven Self-Organizing Fuzzy Neural Network.

Fuzzy neural networks (FNNs) hold the advantages of knowledge leveraging and adaptive learning, which have been widely used in nonlinear system modeling. However, it is difficult for FNNs to obtain the appropriate structure in the situation of insufficient data, which limits its generalization performance. To solve this problem, a data-knowledge-driven self-organizing FNN (DK-SOFNN) with a structure compensation strategy and a parameter reinforcement mechanism is proposed in this article. First, a structure compensation strategy is proposed to mine structural information from empirical knowledge to learn the structure of DK-SOFNN. Then, a complete model structure can be acquired by sufficient structural information. Second, a parameter reinforcement mechanism is developed to determine the parameter evolution direction of DK-SOFNN that is most suitable for the current model structure. Then, a robust model can be obtained by the interaction between parameters and dynamic structure. Finally, the proposed DK-SOFNN is theoretically analyzed on the fixed structure case and dynamic structure case. Then, the convergence conditions can be obtained to guide practical applications. The merits of DK-SOFNN are demonstrated by some benchmark problems and industrial applications.

Quantification of toxic metals in chicken egg and chicken feed via SOM-artificial neural network.

Poultry products such as meat and eggs are rich sources of proteins, vitamins, and minerals. It is a good indicator of healthy food. Keeping in view, the present study is designed to evaluate the prevalence of toxic heavy metals (lead, nickel, cadmium, and chromium) in chicken eggs and feed. For this purpose, five samples of egg and feed were collected from five different commercial markets in Skardu City. Each sample was prepared using the wet digestion method and analyzed using a flame atomic absorption spectrophotometer. The results showed that lead, nickel, and chromium were present in varying amounts in the feed and egg, with nickel being the most concentrated metal, followed by lead and chromium in egg samples, while the feed samples showed the highest concentration of chromium followed by lead and nickel. However, concentrations of selected heavy metals except cadmium were all above the permissible limit of the World Health Organization. The self-organizing map-artificial neural network is employed for the identification of patterns of heavy metals in chicken feed and egg samples. The lower left neurons of the maps showed higher heavy metal concentrations found in samples taken from Bazar, whereas the rest of the samples showed varied concentrations. A comparison of feed and egg concentrations showed that nickel concentration was lower in feed samples than in egg samples. The lead concentration decreased in eggs except in the Krasmathang feed sample. Chromium concentration presented a negative correlation due to the extremely high concentration found in the Bazar feed sample.

Design and prediction of self-organizing interval type-2 fuzzy wavelet neural network

This paper presents a self-organizing interval type-2 fuzzy wavelet neural network (SIT2FWNN) model for predicting and identifying nonlinear systems. Based on traditional TSK interval type-2 fuzzy neural network (IT2FNN), the proposed SIT2FWNN utilizes the wavelet basis function with the locating abilities of time-domain and frequency-domain as the consequent of fuzzy rules, combining the capacity of IT2FNN to handle the uncertainty and the great learning potentiality of wavelet neural network (WNN). For the structural adjustment of SIT2FWNN, fuzzy rules are added and deleted according to the Euclidean distance between the input layer and the fuzzification layer. The self-organizing algorithm can delete redundant and unimportant fuzzy rules, thus optimizing the structure of SIT2FWNN and simplifying the calculation. In parameter learning, the AdaBound algorithm and self-adaptive gradient learning algorithm are used to find optimal values of unknown parameters of the SIT2FWNN model. Finally, the designed SIT2FWNN model is applied in the predictions of short-term traffic flow, Mackey-Glass time series, and the opening index of the Shanghai stock index. The evaluation comparison between the proposed model and similar studies proves that SIT2FWNN has higher prediction accuracy and speed.

Experimental investigation on acoustic emission precursor of rockburst based on unsupervised machine learning method

The key to achieving rockburst warning lies in the understanding of rockburst precursors. Considering the correlation characteristics of rockburst acoustic emission (AE) parameters, a self-organizing map neural network (SOMNN) based method for rockburst precursor inversion was proposed. The feature of this method lies in a cyclic data segmentation iteration process based on the thinking of “interference signal screening”, “key signal extraction”, and “precursor signal inversion”. The rationality of this method has been verified in three groups of rockburst experiments. The results revealed that rockburst AE precursor signals consist of a series of signals characterized by long duration, high energy, low average frequency, high energy amplitude, and low peak frequency. Subsequently, potential value in long term rockburst warning of the precursor obtained in this study was shown via the comparison of conventional precursors. Finally, a preliminary interpretation for rockburst precursor was proposed under the framework of AE parameters physical significance, and it is revealed that AE precursor signals are likely linked to the creation of large-scale tensile cracks before rockburst.

Approximate Optimal Adaptive Prescribed Performance Fault-Tolerant Control for Autonomous Underwater Vehicle Based on Self-Organizing Neural Networks

Approximate Optimal Adaptive Prescribed Performance Fault-Tolerant Control for Autonomous Underwater Vehicle Based on Self-Organizing Neural Networks

Self-Organizing Mapping Neural Network Implementation Based on 3-D NAND Flash for Competitive Learning

Self-Organizing Mapping Neural Network Implementation Based on 3-D NAND Flash for Competitive Learning

One network fits all: A self-organizing fuzzy neural network based explicit predictive control method for multimode process

One network fits all: A self-organizing fuzzy neural network based explicit predictive control method for multimode process

Unbiased kidney-centric molecular categorization of chronic kidney disease as a step towards precision medicine

Current classification of chronic kidney disease (CKD) into stages using indirect systemic measures (estimated glomerular filtration rate (eGFR) and albuminuria) is agnostic to the heterogeneity of underlying molecular processes in the kidney thereby limiting precision medicine approaches. To generate a novel CKD categorization that directly reflects within kidney disease drivers we analyzed publicly available transcriptomic data from kidney biopsy tissue. A Self-Organizing Maps unsupervised artificial neural network machine-learning algorithm was used to stratify a total of 369 patients with CKD and 46 living kidney donors as healthy controls. Unbiased stratification of the discovery cohort resulted in identification of four novel molecular categories of disease termed CKD-Blue, CKD-Gold, CKD-Olive, CKD-Plum that were replicated in independent CKD and diabetic kidney disease datasets and can be further tested on any external data at kidneyclass.org. Each molecular category spanned across CKD stages and histopathological diagnoses and represented transcriptional activation of distinct biological pathways. Disease progression rates were highly significantly different between the molecular categories. CKD-Gold displayed rapid progression, with significant eGFR-adjusted Cox regression hazard ratio of 5.6 [1.01-31.3] for kidney failure and hazard ratio of 4.7 [1.3-16.5] for composite of kidney failure or a 40% or more eGFR decline. Urine proteomics revealed distinct patterns between the molecular categories, and a 25-protein signature was identified to distinguish CKD-Gold from other molecular categories. Thus, patient stratification based on kidney tissue omics offers a gateway to non-invasive biomarker-driven categorization and the potential for future clinical implementation, as a key step towards precision medicine in CKD.