Competitive Learning Research Articles

Unsupervised neural network-enabled spatial-temporal analytics for data authenticity under environmental smart reporting system

Environment, social and governance (ESG) disclosures required on listed companies have aroused considerable interest in both academia and industry for sustainable development and investing. However, the authenticity and credibility of the ESG report exposed to the public remain in doubt due to black box-like reporting processes with massive human involvements. In this study, a framework of environmental smart reporting system (BI-ESRS) based on the blockchain and Internet of Things (IoT) technologies is developed to automate the acquisition of environment-related data and make the reporting reliable and traceable. In addition, we evaluate the authenticity of the data collected from IoT devices, considering human-made counterfeits on measuring instruments for greenwashing. It is anticipated to stimulate companies to submit high-quality data without fake. Specifically, an unsupervised neural network-enabled spatial-temporal analytics (UN-STA) method is devised to achieve anomalies detection and index the data with an authenticity rate. An artificial neural network, self-organizing map (SOM), is applied to construct the prediction model. The received signal strength indicator (RSSI) of Bluetooth low energy (BLE), the vibration amplitude of smart instruments and data uploading interval constitute the input vector for competitive learning. Finally, an experimental simulation is carried out to demonstrate the implementation of the proposed system and method, and their effectiveness has also been testified. Moreover, the sensitivity of the SOM model over the three factors has been analyzed by applying the control variate technique. This work is expected to serve as a reference for practitioners to satisfy similar requirements in the industry and inspire new ideas for scholars.

Data imputation in deep neural network to enhance breast cancer detection

AbstractBreast cancer is one of the most precarious cancers that claims many women's' lives every year. The existing automated systems for mammography datasets are designed to detect the abnormalities and classify them as benign or malignant. However, these systems ignore the incompleteness of data due to missing or irrelevant values in the dataset, which severely impede the automated system's reliability. To enhance the reliability of the decision outcomes, this study proposes a composite imputed multilayer fully connected deep networks (MFCN) that can replace the incomplete values in the dataset to augment the efficient feature learning for breast cancer classification. The data imputation techniques and one‐hot encoding transformation applied to the dataset are proved to improve the feature representation. The imputed features extracted from the instances are fed into the MFCN to classify malignant and benign by utilizing mammographic mass (MM) and INbreast datasets. The performance of the proposed missing data imputation treated MFCN (MdI‐MFCN) is compared with competitive machine learning (ML) classifiers by incorporating stratified 6‐fold cross‐validation. The experimental results of the models are evaluated with and without imputed data using the five performance metrics, namely, accuracy, precision, recall, F1‐score, and specificity. The performance of the proposed MdI‐MFCN has outperformed the competitive classifiers with the highest accuracy of 93% and 98% of the area under the receiver operating characteristic (ROC) curve on MM and INbreast datasets, respectively. The simplicity and reliability of this proposed approach its potential in characterizing mammography datasets during breast cancer diagnosis.

The Concurrent and Longitudinal Relations Between Competitive Classroom Climate and Learning Motivation Among Chinese Adolescent Students: The Mediating Roles of Social Comparisons.

PurposeOver the past decade, cross-sectional studies have established a link between competitive classroom climate and learning motivation. However, the precise predictive direction has remained unclear, and the potential mechanisms underlying the link have yet to be investigated. According to the social comparison theory, competitive classroom climate is positively associated with learning motivation, and upward and downward comparison may play a role in mediating this process.Patients and MethodsData were obtained from a three-wave study of 476 Chinese senior high school students (45.4% boys, 54.6% girls) aged 14–18 years. Structural equation modeling was performed to test different models.ResultsConcurrent and longitudinal analyses revealed that higher competitive classroom climate was associated with higher learning motivation. Moreover, upward comparison mediated the relation between competitive classroom climate and learning motivation. However, downward comparison did not mediate these relations.ConclusionFrom the perspective of practice, the results provide evidence for how to promote learning motivation by cultivating competitive classroom climate and applying effective social comparison strategies.

Competitive Learning of Facial Fitting and Synthesis Using UV Energy

The three-dimensional morphable model (3DMM) is the most widely used representative model for obtaining a three-dimensional (3-D) face from a target on an image. Although 3DMMs have demonstrated the powerful capability to represent various facial shapes on natural images, they are limited to capturing texture variations of in-the-wild human faces. Based on the fact that fitting a 3-D facial model to an image determines the corresponding UV map, we propose a novel method for facial fitting and synthesis by competitively training two deep learning networks for facial alignment and UV texture completion. When the completion network is trained using well-aligned UV maps, it can model facial textures precisely and, consequently, fill the missing regions more completely. Accordingly, we use a UV completion network, denoted as a UV energy-based generative adversarial network (UV EB-GAN), to discriminate whether a UV map from the alignment network is well aligned by defining the generative loss of the completion network as the energy. Competitive learning facilitates training the completion network without ground-truth facial UV maps and training the alignment network without hard constraints and regularization terms. The proposed network can be trained in an end-to-end manner. The facial texture, albedo, lighting parameters, and 3-D facial shape can be obtained through this network. The results of the experiments on 2-D alignment, 3-D reconstruction, texture synthesis, and illumination estimation verified that the proposed method achieves remarkable improvements over the state-of-the-art methods.

CPRO: Competitive Poor and Rich Optimizer-Enabled Deep Learning Model and Holoentropy Weighted-Power K-Means Clustering for Brain Tumor Classification Using MRI

A brain tumor is a collection of irregular and needless cell development in the brain region, and it is considered a life-threatening disease. Therefore, early level segmentation and brain tumor detection with Magnetic Resonance Imaging (MRI) is more important to save the patient’s life. Moreover, MRI is more effective in identifying patients with brain tumors since the recognition of this modality is moderately larger than considering other imaging modalities. The classification of brain tumors is the most important, difficult task in medical imaging systems because of size, appearance and shape variations. In this paper, Competitive Poor and Rich Optimization (CPRO)-based Deep Quantum Neural Network (Deep QNN) is proposed for brain tumor classification. Additionally, the pre-processing process assists in eradicating noises and uses image intensity to eliminate the artifacts. The significant features are extracted from pre-processed image to perform a productive classification process. The Deep QNN classifier is employed for classifying the brain tumor regions. Besides, the Deep QNN classifier is trained by the developed CPRO approach, which is newly designed by integrating Poor and Rich Optimization (PRO) and Competitive Swarm Optimizer (CSO). The developed brain tumor detection model outperformed other existing models with accuracy, sensitivity and specificity of 94.44%, 97.60% and 93.78%.

Partial Label Learning with competitive learning graph neural network

Partial Label Learning (PLL) is a weakly supervised learning framework where each instance may be associated with more than one candidate label, among which only one is true. Traditionally, the PLL problem is solved by removing the false candidate labels based on the instance relationship, while the potentially useful information between instances and labels as well as the potential candidate label relationship is ignored. In this paper, a new PLL framework PL-CGNN is proposed, which treats the instances with false labels as noise, and the PLL is reformulated to remove the noise instances. First of all, the feature of each label class is approximately represented by the center point of all the related instances. The significant operation enables the similarity between instances and labels measurable. Next, all the candidate labels for each instance compete for the biggest similarity. To further improve the robustness of the model, the competition procedure for the most similar label is extended to the neighbors of this instance. The label with the most wins is the final ground-truth one. The relationship between candidate labels guides the situation that the competition process develops into. Through iterative competitive learning, each label class approaches the true value. Experiments carried out on diverse datasets show that the performance of the PL-CGNN model is outstanding.

Changes in urban air pollution after a shift in anthropogenic activity analysed with ensemble learning, competitive learning and unsupervised clustering

Urban air pollution is a health hazard linked to anthropogenic emissions. Reliable evaluation of changes in pollutants due to altered emissions requires considering meteorological and other variability influencing concentrations. Here, a combination of ensemble learning, competitive learning and unsupervised clustering is proposed and applied to leverage the change analysis of particulate matter (PM2.5) and other pollutants.Machine Learning (ML) algorithms Random Forest (RF) and Self-Organizing Map (SOM) were trained with historical meteorological data, pollutant concentrations and traffic indicators. The importance of different variables for local PM2.5 was determined with RF. SOM was configured for multivariable cluster analysis. The trained SOM enabled predicting a cluster for new data representing conditions with shifted anthropogenic activity. The prediction forms a benchmark for the analysed period with maximized meteorological similarity, which facilitates identifying changes in ambient pollutants due to changed emissions.The method was applied to data from the start of COVID-19 pandemic, 3/2020, when emissions suddenly decreased. For measurements from Helsinki, Finland, the SOM yielded a statistically significant change in PM2.5 (−0.7%), NO2 (−33%) and O3 (+17%). Comparing data from 3/2020 to data from 3/2017–2019 produced different results (PM2.5 −1.7%, NO2 −37%, O3 −4.0%). Statistical indicators confirmed better compatibility between the analysed period and its benchmark when using the SOM prediction instead of calendar-based selection: Average RMSRE was 19%-points lower and Willmott's dr 41% higher with SOM than with 3/2017–2019.Based on the case study and method evaluation, using ML for multivariate analysis of changed air pollution is feasible and yields meaningful results.

Quiz as an academic tool for teaching learning physiology in Indian medical students: A cross sectional study

Background: In a changing World, education has become a priority. It has become need of the hour to explore more innovative ways of teaching effectively. Active teaching in the form of small group discussions and interactive activities like quiz are been explored to ensure healthy and competitive learning. Not many studies have been done in India and thus, this study was undertaken to assess the effectiveness of quiz as an academic tool. Methods: 96 First year medical undergraduates participated in quizzes on various topics of Physiology that were organized every 3-4 weeks. At the end of the year, they were asked to give their feedback anonymously in a Likert scale that was duly analyzed. Results:49% students strongly liked the quiz, 40% liked it and only 1% of the students did not like it. Majority of the students liked various contents and organization of the quiz. Most of the students liked the procedure of the quiz e.g. pattern of team formation, weightage of topics, frequency, duration, pattern of scoring, time for answering questions, contents, rounds, difficulty level and usefulness in study. Conclusion: Activities like quizzes are beneficial to the students and must be encouraged as a part of the curriculum.

Effects of Online Learning on College Students in Eastern China: A Structural Equation Model.

With the spread of COVID-19 worldwide, online education is rapidly catching on, even in some underdeveloped countries and regions. Based on Bandura's ternary learning theory and literature review, this paper takes online learning of college students as the research object and conducts an empirical survey on 6,000 college students in East China. Based on literature review and factor analysis and structural equation model, this paper discusses the relationship among learning cognition, learning behavior, and learning environment in online learning of college students. The learning behavior includes interactive communication, self-discipline mechanism, classroom learning, and study after class. The learning environment includes teaching ability, knowledge system, platform support, process control, and result evaluation; learning cognition includes learning motivation, information perception, and adaptability. It is found that the learning environment has a significant positive impact on learning behavior, and learning cognition has a significant positive impact on learning behavior. It is uncertain whether the learning environment significantly impacts learning cognition. At the learning environment level, the teaching ability (0.59) has the most significant impact on the learning environment, followed by result evaluation (0.42), platform support (0.40), process control (0.33), and knowledge system (0.13). In terms of the influence on learning behavior, classroom learning has the most significant impact (0.79), followed by self-discipline mechanism (0.65), study after class (0.54), and interactive communication (0.44). In terms of learning cognition, information perception (0.62) has the most significant influence, followed by learning motivation (0.50) and adaptability (0.41). This paper suggests strengthening the construction of platforms and digital resources to create a more competitive learning environment. Improve process management and personalized online services, constantly stimulate students' enthusiasm for independent online learning, and cultivate students' online independent learning ability to promote the sustainable and healthy development of online education.

Prediction of misfire location for SI engine by unsupervised vibration algorithm

The early detection of faults by monitoring can prevent further damage to the internal combustion engine and avoid further causalities. The faults lead to reducing the engine performance and increasing the harmfulpollution. In this paper, a novel unsupervised vibration-based detection algorithm for predicting the location of the misfire of the spark-ignited engine using Khonen self-organizing map (SOM) is introduced. The purpose is to map the misfire location on self-organizing neural network with the best quality comparing 1D, 2D and 2D vibration signals. Four fault locations of the 4-cylinder engine are clustered and mapped at normal and misfire conditions. The behavior of the SOM for the engine vibration signals is investigated. The clustering neural network with a competitive learning algorithm is used to decide which cylinder has the misfire. The output map is used to cluster misfires using a tri-axial accelerometer to record vibration signals for three operation conditions. Also, the mean quantization error is used to measure the quality of each map. It is found that the ability of the SOM to discern misfire locations at three-dimensional vibration signals. Finally, it is found that the 3D vibration signal is improved the clustering with minimum errors 0.0025, 0.0011, 0.00043 for three different operation conditions. In additions, the maximum accuracy in detecting the misfire is 93.55% for 3D vibration signal at 3000 rpm.

Construction of Dynamic Multiparallel Foreign Language Teaching Model Based on Multicore Processor

Heterogeneous multicore processor systems, as one of the highlights of multicore processor systems, are widely loved by people for their high efficiency and low cost, and they have also become the most commonly used processor systems in embedded systems. In the process of research on heterogeneous multicore processor systems, system task scheduling is particularly important. A good task scheduling algorithm can give full play to system performance. In this paper, the intelligent approximation algorithm is applied to the task scheduling problem of heterogeneous multicore processor system, and the heterogeneous multicore processor system is obtained as a highlight in the multicore processor system. Relying on the characteristics of high efficiency and low cost, it is widely loved by people, and at the same time, it has become the most commonly used processor system in embedded systems. In the process of research on heterogeneous multicore processor systems, the system task scheduling problem is particularly important. A good task scheduling algorithm can give full play to system performance. Some commonly used heuristic task scheduling algorithms are insufficient in solving such problems. This paper combines the granularity-based wavefront parallel decoding algorithm and the fast fusion loop filter algorithm to apply pipeline parallel technology between pixels. Decoding reconstruction module and fast loop filter module realize the fusion of multilevel parallel decoding. Based on the multicore platform, a dynamic multiparallel scheduling algorithm is designed to realize two-way video real-time parallel high-speed decoding, which improves the core resource utilization rate and decoding execution efficiency of the multicore processing platform. This paper also designs an indicator collector, a read-write hit collector, and a cache block priority determiner to implement a dynamic generation strategy with low hardware overhead. Multimodal teaching has certain feasibility and effectiveness and can have a positive impact on the English reading motivation and English reading comprehension ability of university students. Multimodal teaching improves students’ English reading comprehension ability, deepens students’ understanding and memory of words, and broadens the scope of knowledge, which has a significant promoting effect. The research results show that multimodality can be applied to college English teaching, and it can achieve better results than traditional teaching methods. By comparing the test results before and after the test, there are obvious differences between the experimental class and the control class. Whether it is the paired sample T-test of the experimental class or the independent sample T-test of two teaching methods in two classes, it proves that the scores under the multimodal English teaching mode are higher than the traditional teaching mode. Multimodal classrooms provide students with many opportunities to participate in classroom activities and form a competitive learning atmosphere. This competitive learning atmosphere has become a driving force to promote student learning. Multimodal teaching methods help to cultivate students’ independent and cooperative learning. In a multimodal classroom, many activities require cooperation and discussion among students. When they encounter difficulties, they can help each other, discuss with each other, and cooperate to complete tasks. Therefore, the multimodal teaching method will stimulate students’ interest in learning, give full play to students’ initiative, and improve students’ English ability.

Affective Valence Regulates Associative Competition in Pavlovian Conditioning.

Evidence suggests that, in Pavlovian conditioning, associations form between conditioned stimuli and multiple components of the unconditioned stimulus (US). It is common, for example, to regard USs as composed of sensory and affective components, the latter being either appetitive (e.g., food or water) or aversive (e.g., shock or illness) and, therefore, to suppose different USs of the same affective class activate a common affective system. Furthermore, evidence is growing for the suggestion that, in competitive learning situations, competition between predictive stimuli is primarily for association with the affective system activated by the US. Thus, a conditioned stimulus (CS) previously paired with one US will block conditioning to another CS when both are presented together and paired with a different US of the same affective class, a phenomenon called transreinforcer blocking. Importantly, similar effects have been reported when steps are taken to turn the pretrained CS into a conditioned inhibitor, which activates the opposing affective state to the excitor that it inhibits. Thus, an appetitive inhibitor can block conditioning to a second CS when they are presented together and paired with foot shock. Here we show that the same is true of an aversive inhibitor. In two experiments conducted in rats, we found evidence that an aversive inhibitor blocked conditioning to a second CS when presented in a compound and paired with food. Such findings demonstrate that affective processes and their opponency organize appetitive-aversive interactions and establish the valences on which they are based, consistent with incentive theories of Pavlovian conditioning.

Hawk and pigeon’s intelligence for UAV swarm dynamic combat game via competitive learning pigeon-inspired optimization

Hawk and pigeon’s intelligence for UAV swarm dynamic combat game via competitive learning pigeon-inspired optimization

Unsupervised Digit Recognition Using Cosine Similarity In A Neuromemristive Competitive Learning System

This work addresses how to naturally adopt the l 2 -norm cosine similarity in the neuromemristive system and studies the unsupervised learning performance on handwritten digit image recognition. Proposed architecture is a two-layer fully connected neural network with a hard winner-take-all (WTA) learning module. For input layer, we propose single-spike temporal code that transforms input stimuli into the set of single spikes with different latencies and voltage levels. For a synapse model, we employ a compound memristor where stochastically switching binary-state memristors connected in parallel, which offers a reliable and scalable multi-state solution for synaptic weight storage. Hardware-friendly synaptic adaptation mechanism is proposed to realize spike-timing-dependent plasticity learning. Input spikes are sent out through those memristive synapses to each and every integrate-and-fire neuron in the fully connected output layer, where the hard WTA network motif introduces the competition based on cosine similarity for the given input stimuli. Finally, we present 92.64% accuracy performance on unsupervised digit recognition with only single-epoch MNIST dataset training via high-level simulations, including extensive analysis on the impact of system parameters.

COMPETITIVE LEARNING IN NEURAL NETWORKS

The article presents the basic concept of competitive learning in neural networks. Provides the main machine learning learning models and applications. The analysis of the advantages and disad-vantages of these models is carried out. The geometric interpretation of competitive learning is presented in terms ofmathematical formulas, as well as the behavior of neurons in this model. Neural systems are described as a powerful tool and driver in the field of modern Internet technologies, in data science and big (meta) data.

Platform Leadership and Sustainable Competitive Advantage: The Mediating Role of Ambidextrous Learning.

In the context of the knowledge economy, the role of traditional leadership for enterprises is questioned. Based on contingency theory and the resource-based view, this paper proposes the important role of platform leadership, a new leadership type in line with the context of the times, for a sustainable competitive advantage. We conducted an empirical study to examine and confirm the positive effects of platform leadership on sustainable competitive advantage and ambidextrous learning. We also verified the mediation effect of exploratory and exploitative learning on platform leadership and sustainable competitive advantage. Additionally, relevant discussion and research contributions are put forward.

Multi-Agent Reinforcement Learning for Joint Cooperative Spectrum Sensing and Channel Access in Cognitive UAV Networks.

This paper studies the problem of distributed spectrum/channel access for cognitive radio-enabled unmanned aerial vehicles (CUAVs) that overlay upon primary channels. Under the framework of cooperative spectrum sensing and opportunistic transmission, a one-shot optimization problem for channel allocation, aiming to maximize the expected cumulative weighted reward of multiple CUAVs, is formulated. To handle the uncertainty due to the lack of prior knowledge about the primary user activities as well as the lack of the channel-access coordinator, the original problem is cast into a competition and cooperation hybrid multi-agent reinforcement learning (CCH-MARL) problem in the framework of Markov game (MG). Then, a value-iteration-based RL algorithm, which features upper confidence bound-Hoeffding (UCB-H) strategy searching, is proposed by treating each CUAV as an independent learner (IL). To address the curse of dimensionality, the UCB-H strategy is further extended with a double deep Q-network (DDQN). Numerical simulations show that the proposed algorithms are able to efficiently converge to stable strategies, and significantly improve the network performance when compared with the benchmark algorithms such as the vanilla Q-learning and DDQN algorithms.

Morphology-aware multi-source fusion-based intracranial aneurysms rupture prediction.

We proposed a new approach to train deep learning model for aneurysm rupture prediction which only uses a limited amount of labeled data. Using segmented aneurysm mask as input, a backbone model was pretrained using a self-supervised method to learn deep embeddings of aneurysm morphology from 947 unlabeled cases of angiographic images. Subsequently, the backbone model was finetuned using 120 labeled cases with known rupture status. Clinical information was integrated with deep embeddings to further improve prediction performance. The proposed model was compared with radiomics and conventional morphology models in prediction performance. An assistive diagnosis system was also developed based on the model and was tested with five neurosurgeons. Our method achieved an area under the receiver operating characteristic curve (AUC) of 0.823, outperforming deep learning model trained from scratch (0.787). By integrating with clinical information, the proposed model's performance was further improved to AUC = 0.853, making the results significantly better than model based on radiomics (AUC = 0.805, p = 0.007) or model based on conventional morphology parameters (AUC = 0.766, p = 0.001). Our model also achieved the highest sensitivity, PPV, NPV, and accuracy among the others. Neurosurgeons' prediction performance was improved from AUC=0.877 to 0.945 (p = 0.037) with the assistive diagnosis system. Our proposed method could develop competitive deep learning model for rupture prediction using only a limited amount of data. The assistive diagnosis system could be useful for neurosurgeons to predict rupture. • A self-supervised learning method was proposed to mitigate the data-hungry issue of deep learning, enabling training deep neural network with a limited amount of data. • Using the proposed method, deep embeddings were extracted to represent intracranial aneurysm morphology. Prediction model based on deep embeddings was significantly better than conventional morphology model and radiomics model. • An assistive diagnosis system was developed using deep embeddings for case-based reasoning, which was shown to significantly improve neurosurgeons' performance to predict rupture.

The Impact of Collaborative Learning Strategies and Learning Motivated Students on Learning outcomes and Ability in Madrasah Ibtidaiyah

There will be an examination of the impact of cooperative learning strategies on academic outcomes for students of Akidah Aklak at Madrasah Ibtidayah Negeri Gunungsitoli City, in this study. This study used a quasi-experimental approach. For both collaborative and competitive learning classes, a previously formed class was used for treatment, and the controlled characteristics of students were learning motivation. This method was chosen. Using a variety of collaborative learning strategies, this study found that both students with high and low levels of learning motivation at the State Madrasah Ibtidiyah in Gunungsitoli City could improve their academic performance. To teach Akidah Akhlak material, collaborative learning strategies should be used rather than competitive learning strategies.

The effect of perceptions competition and learning costs on cooperation in spatial evolutionary multigames

It is a great challenge to correctly understand the evolution and incentives of cooperation. In recent years, the spatial evolutionary multigames have attracted extensive attention from researchers. Actually, for players, not only themselves but also their neighbors will affect their perceptions, which has not been considered in previous works. In addition, changing perceptions has a learning process, which will cause some costs. Inspired by the above facts, we explore the impact of perceptions competition mechanisms and learning costs on the evolution of cooperation in the multigames dilemma, which consists of prisoner game and snowdrift game. Especially, we further explore how different groups can coexist in such a setting. It is found that the heterogeneity of payoffs contributes to network reciprocity and can maintain cooperative behaviors even under unfavorable conditions. Compared with the static multigames mechanism, our perceptions competition mechanism has a higher proportion of cooperators under the same temptation value b and sucker payoff θ, and the cooperators can survive within a larger range of parameters. At the same time, a small cost can effectively maintain a high proportion of cooperators, while an appropriate cost will keep players at a high level of both the proportion of cooperators and the coexistence range. The continuous increase of costs will make our evolutionary mechanism ineffective and turn to static multigames. We also find that the introduction of costs will help different types of players coexist, and with the increase of costs, the coexistence area will first decrease and then increase. Moreover, our main findings are equally applicable to complex networks such as BA networks and random networks.