Waterway transportation sustainably facilitates global trade through eco-efficient cargo movement, where accurate water level forecasting is critical for ensuring navigational safety and operational continuity. To develop a highly accurate prediction model, it is essential to consider the periodic characteristics of water level data, which often emerge in real-world datasets. This study introduces a novel reservoir state structure based on reservoir computing theory, the Self-attention Hierarchical Kernel Reservoir State Network (SHK-RSN). It employs three primary mechanisms. First, a hierarchical feature extraction method groups training data and extracts high-dimensional features from these groups using the kernel trick in a hierarchical manner. Second, a self-attention weight selection approach is introduced to replace the random weights in the Hierarchical Kernel Reservoir State Network (HK-RSN), improving the rationale for hidden neuron connections and enhancing the interpretability of weight selection. Third, a novel reservoir state structure is proposed to capture periodic information and extract temporal features across periods, enabling the model to capture richer temporal information and identify relationships among periods. Experiments are conducted on one artificial and five real-world time series datasets, with forecast performance evaluated over 1–7 steps. Our proposed model, SHK-RSN, is compared with models based on randomization, the kernel trick, and deep learning. The experimental results demonstrate that SHK-RSN exhibits superior forecasting ability relative to the baselines. It achieves the best Symmetric Mean Absolute Percentage Error (SMAPE) across all datasets in the 1–7 period average among baseline methods, demonstrating a relative improvement of 25.7% to 46.9% over the conventional Echo State Network.

Knowledge probing plays a crucial role in evaluating the factual knowledge retention and reasoning capabilities of large language models (LLMs), particularly in high-risk biomedical applications. However, existing biomedical probing benchmarks often fail to capture the complexity of real-world medical reasoning. In this study, we introduce MedLAMA2, an advanced biomedical knowledge probing benchmark built upon MedLAMA, incorporating multi-token entities and two-hop reasoning queries. All vocabularies are sourced from the Unified Medical Language System (UMLS) Metathesaurus**, ensuring domain-specific accuracy.To systematically evaluate LLMs without fine-tuning, we design a structured probing pipeline using Top-k Metric and Token-level Metric as evaluation parameters. We assess the performance of encoder-only and decoder-only models across different parameter scales, ranging from 6B to 72B. Our experiments demonstrate that decoder-only models outperform encoder-only models, particularly in complex multi-hop reasoning tasks. However, even the best-performing models, such as Llama3-70B-Instruct and Qwen-1.5-72B-Chat, achieve acc@10 scores below 40%, highlighting the limitations of current LLMs in structured biomedical knowledge retrieval. Additionally, our results show that Chain-of-Thought (CoT) prompting does not significantly improve the accuracy of biomedical knowledge probing tasks and may even degrade performance in multi-hop settings. This suggests that CoT may not be well-suited for structured medical knowledge retrieval.This study highlights the challenges of biomedical knowledge probing and aims to construct a new benchmark and evaluation process to provide more effective methods for assessing LLM performance in real-world medical environments, ultimately promoting the application of LLMs in the biomedical domain.

The combination of wearable sensors and competitive sports provides quantitative information for scientific training, effectively assisting athletes in improving their athletic performance. This study presents a technical framework for athletic sports assessment in competitive swimming based on body-area sensor networks. In our approach, wearable inertial sensor nodes are placed on specific body parts of the athletes to capture motion data during different competitive swimming strokes. Multiwearable inertial sensor nodes are worn on specific body parts of athletes for real-time monitoring motion data during training sessions. A motion intensity detection-based error-state-Kalman-filter algorithm is proposed for multisensor data fusion. Additionally, through kinematic statistical analysis, the characteristics of joint motion during training are clearly explained. Furthermore, a deep learning network that fuses sensor time series and human skeleton graphs is proposed for different stroke phase segmentation, enabling quantitative measurement of motion phases, and several baseline classifiers are chosen for comparison to validate the robustness of our phase segmentation method. We also investigate the sensor combination selection issue during the phase segmentation process to determine the optimal sensor configuration. Our approach provides a scientific solution for the integration of wearable sensors and competitive sports, contributing to the high-quality development of the next generation of smart sports.

Gait phase segmentation, pivotal for understanding lower limb motion, finds applications in diverse fields like medicine and sports. While existing method often struggle with accuracy and adaptability in real-world settings, this study presents a novel methodology employing particle filters for precise lower limb motion capture (MoCap) utilizing inertial sensors, which can be used in more everyday environments and in a wider range of applications over a longer period of time. The innovative approach adeptly tracks walking movements, labeling six gait phases via skeleton reconstruction facilitated by the MoCap algorithm. Subsequently, we propose a neural network architecture amalgamating temporal convolutional network (TCN), graph convolutional network (GCN), and long short-term memory (LSTM). This architecture integrates raw data from inertial sensors with joint angles derived from reconstructed motion, achieving accurate segmentation of the six gait phases. Experimental validation compares the MoCap algorithm against an optical motion capture system, and the neural network’s performance against state-of-the-art methods. Results demonstrate our method’s superior accuracy of 96.94%, highlighting its efficacy in addressing gait phase segmentation challenges and propelling advancements in gait analysis.

Utilizing artificial intelligence techniques for automated diagnosis of cancerous areas within gastric tissue pathology images can significantly augment physicians' diagnostic capabilities and subsequent treatment procedures. However, conventional segmentation models based on UNet architecture typically have images of the lesion area as outputs, failing to reconstruct the original gastric tissue pathology images from learned image features. This approach often results in incomplete learning of the complex details within gastric tissue pathology images, rendering the learned features susceptible to noise interference. Furthermore, previous segmentation models have used pooling operations, such as max, random, or average pooling, neglecting the holistic and global features present in gastric tissue pathology images, consequently failing to represent pathological features of gastric cancer tissue effectively. Therefore, we propose a reconstructed UNet model with hybrid fuzzy pooling (RUHFP) to detect lesion areas within gastric tissue pathology images. The RUHFP model is primarily based on the UNet architecture. The UNet version used in this work is U2-Net. Its novelty lies in integrating reconstruction operations from autoencoders into the UNet architecture. We jointly optimize the loss functions of both decoders to enhance the robustness of learned image features against noise interference. In addition, we incorporate fuzzy pooling operations for feature extraction, which are fused with features learned by the UNet architecture to improve the effectiveness and interpretability of image features. Several experimental tests conducted on real gastric tissue pathology image datasets validate the outstanding performance of the RUHFP model.

Combining pre-trained language models (PLMs) and manual templates is a common practice for text classification in zero-shot scenarios. However, the effect of this approach is highly volatile, ranging from random guesses to near state-of-the-art results, depending on the quality of the manual templates. In this paper, we show that this instability stems from the fact that language models tend toward predicting certain label words of text classification, and manual templates can influence this tendency. To address this, we develop a novel pipeline for annotating and filtering a few examples from unlabeled examples. Moreover, we propose a new method to measure model bias on label words that utilizes unlabeled examples as a validation set when tuning language models. Our approach does not require any pre-labeled examples. Experimental results on six text classification tasks demonstrate that the proposed approach significantly outperforms standard prompt learning in zero-shot settings, achieving up to 19.7% absolute improvement and 13.8% average improvement. More surprisingly, on IMDB and SST-2, our approach even exceeds all few-shot baselines.

CiRA CORE is a central hub designed to connect AI technology creation with practical application, making it easier to work with ROS (Robot Operating System) and link different systems through a user-friendly drag-and-drop interface. This approach removes the need for extensive coding, making the platform accessible to those with minimal programming experience. CiRA CORE offers a comprehensive suite of features for AI development and robot control, including algorithm creation, AI model training, and device integration commonly used in industrial settings. It supports tasks like image recognition and facilitates data storage, labeling, and integration with other systems for data-driven AI development. Overall, CiRA CORE aims to democratize AI development and robot control, simplifying AI development for Industry 4.0 applications, and leading to increased efficiency, reduced costs, and improved safety in industrial processes. This paper reports the progress of the CiRA CORE training modules funded by the SMCS TEAM Program Award. The project has completed the design of a 6-axis robot 3D training kit and simulation models for CiRA CORE training modules. The next steps involve developing 3D-printed robots and training materials. The main goal is to democratize advanced robotics and AI by simplifying integration through a visual, node-based programming interface. This approach reduces the need for complex coding, making these technologies accessible to users with limited programming experience. This initiative aims to foster widespread adoption in business and industrial settings, aligning with IEEE SMC's mission to promote professional growth and innovation in robotics and AI.

This research presents a comprehensive investigation into the dynamics of forested ecosystems using advanced geospatial techniques and machine learning applications, focusing on the University of Malaya study area. The study aims to contribute crucial data for informed decision-making aligned with sustainable development goals. It encompasses canopy height estimation, aboveground biomass density prediction, and carbon stock estimation. Machine learning algorithms, including Random Forest, Gradient Boost Tree Regression, and Support Vector Machine, are employed for canopy height estimation. Their performance is evaluated with and without Principal Component Analysis using metrics such as Root Mean Squared Error and R-squared. Results, summarized in Table 1 and Table 2, highlight the variability in canopy height predictions across different models and feature selection methods. The research explores challenges associated with GEDI Aboveground Biomass Density data, emphasizing spatial variability in model performance across different strata. Results, detailed in Table 3, underscore the importance of tailoring the model, especially in areas characterized by high biomass canopy forests. The integration of Aboveground Biomass Density data with tree cover datasets forms the basis for aboveground carbon stock estimation. Carbon stock is calculated considering forest area, land-use types, and specific carbon content factors. Findings, presented in Table 5, reveal a spectrum of aboveground carbon stock estimates, reflecting the complexity of the University of Malaya study area. This research advances remote sensing and machine learning in forestry and environmental monitoring. Its insights support informed decision-making and policy formulation.

Cognitive load theory suggests that overloading of working memory may negatively affect the performance of human in cognitively demanding tasks. Evaluation of cognitive load is a difficult task; it is often assessed through feedback and evaluation from experts. Cognitive load classification based on Functional Near-InfraRed Spectroscopy (fNIRS) is now one of the key research areas in recent years, due to its resistance of artefacts, cost-effectiveness, and portability. To make fNIRS more practical in various applications, it is necessary to develop robust algorithms that can automatically classify fNIRS signals and less reliant on trained signals. Many of the analytical tools used in cognitive sciences have used Deep Learning (DL) modalities to uncover relevant information for mental workload classification. This review investigates the research questions on the design and overall effectiveness of DL as well as its key characteristics. We have identified 38 studies published between 2011 and 2022, that specifically proposed Machine Learning (ML) models for classifying cognitive load using data obtained from fNIRS devices. Those studies were analyzed based on type of feature selection methods, input, and DL model architectures. Most of the existing cognitive load studies are based on ML algorithms, which follow signal filtration and hand-crafted features. It is observed that hybrid DL architectures that integrate convolution and LSTM operators performed significantly better in comparison with other models. However, DL models especially hybrid models have not been extensively investigated for the classification of cognitive load captured by fNIRS devices. The current trends and challenges are highlighted to provide directions for the development of DL models pertaining to fNIRS research.

In a closed environment lacking global positioning system (GPS) signals, how to achieve accurate navigation and positioning is a very challenging task. Zero velocity update (ZUPT) is a highly effective foot-mounted inertial pedestrian navigation systems in such environment. However, despite its effectiveness, the limitation of accurate detecting the zero-velocity-interval (ZVI) and heading drift are still the significant challenges of the ZUPT method. To address these issues, a deep learning method for adaptive ZVIs detection is established based solely on inertial sensors by comparing with the optical motion capture system. Additionally, an improved ZUPT-aided extend Kalman filter (EKF) divides the measurement updates of the ZVIs is established for multisensor data fusion, and the heading change with heuristic drift reduction (HDR) is also adopt as measurement, thereby yielding to limit the heading drift. Experimental results demonstrate that our method provides a better estimate of the heading angle, as well as more accurate ZVIs detection, leading to more precise dead-reckoning position estimates than other state-of-the-art methods.

Gastric cancer has a high incidence rate, significantly threatening patients' health. Gastric histopathology images can reliably diagnose related diseases. Still, the data volume of histopathology images is too large, making misdiagnosis or missed diagnosis easy. The classification model based on deep learning has made some progress on gastric histopathology images. However, traditional convolutional neural networks (CNNs) generally use pooling operations, which will reduce the spatial resolution of the image, resulting in poor prediction results. The image feature in previous CNN has a poor perception of details. Therefore, we design a dilated CNN with a late fusion strategy (DCNNLFS) for gastric histopathology image classification. The DCNNLFS model utilizes dilated convolutions, enabling it to expand the receptive field. The dilated convolutions can learn the different contextual information by adjusting the dilation rate. The DCNNLFS model uses a late fusion strategy to enhance the classification ability of DCNNLFS. We run related experiments on a gastric histopathology image dataset to verify the excellence of the DCNNLFS model, where the three metrics Precision, Accuracy, and F1-Score are 0.938, 0.935, and 0.959.

The lower extremity exoskeleton, which can sense the neural motion state of the human body and then provide motion assistance, is gradually replacing the traditional wheelchairs and assistive devices, making many patients with disabilities or movement disorders able to regain the walking function. This survey provides a comprehensive review on recent technological advances in lower extremity neurorehabilitation exoskeleton from the perspectives of sensing, gait dynamics, and human–robot collaboration. For each technology category, a detailed comparison among state-of-the-art solutions is provided. The results show that the exoskeleton has been greatly improved in mechanical and learning ability. However, some issues, such as adaptability, safety, and efficiency still restrict the development of exoskeleton technology. To address these problems, the remaining open challenges and future directions to improve intelligence, sensing, gait analysis, trust, efficiency, generalization, and power consumption of exoskeleton are also presented and discussed.

This article proposes a multi-label classification algorithm capable of continual learning by applying an Adaptive Resonance Theory (ART)-based clustering algorithm and the Bayesian approach for label probability computation. The ART-based clustering algorithm adaptively and continually generates prototype nodes corresponding to given data, and the generated nodes are used as classifiers. The label probability computation independently counts the number of label appearances for each class and calculates the Bayesian probabilities. Thus, the label probability computation can cope with an increase in the number of labels. Experimental results with synthetic and real-world multi-label datasets show that the proposed algorithm has competitive classification performance to other well-known algorithms while realizing continual learning.

The general success criterion for an artificial intelligence system is its ability to mimic human brain learning. Throughout a lifetime, the human brain is capable of continual learning. The acquired information is kept, augmented, finetuned, and utilized to complete new tasks in the future. At the moment, machine learning models perform well when given precisely structured, balanced, and homogenized data. However, when several jobs with incremental data are provided, the performance of the majority of these models suffers. Inspired by the Complementary Learning Systems (CLS) theory in neuroscience, episodic-semantic memory-based frameworks have received much attention and research. On the other hand, conventional methods are needed to perform data batch normalization and are sensitive to vigilance hyperparameters across different datasets. This paper proposes a Robust Growing Memory Network (RGMN) that continuously learns incoming data without normalization and is unlikely to be affected by the vigilance hyperparameter. The RGMN is a self-organizing topological network that models human episodic memory, and its network size can grow and shrink in response to data. The long-term memory buffer retains the largest and smallest data values that will use for learning. To evaluate the performance of the proposed method, we conducted comparative experiments on real-world datasets, and results showed that the proposed method outperforms existing memory-based baseline frameworks in terms of accuracy.

Lifelong learning is a long-standing aim for artificial agents that act in dynamic environments in which an agent needs to accumulate knowledge incrementally without forgetting previously learned representations. Contemporary methods for incremental learning from images are predominantly based on frame-based data recorded by conventional shutter cameras. We investigate methods for learning from data produced by event cameras and compare techniques to mitigate forgetting while learning incrementally. We propose a model that is composed of both, feature extraction and incremental learning. The feature extractor is utilized as a self-supervised sparse convolutional neural network that processes event-based data. The incremental learner uses a habituation-based method that works in tandem with other existing techniques. Our experimental results show that the combination of different existing techniques with our proposed habituation-based method can help avoid catastrophic forgetting even more, while learning incrementally from the features provided by the extraction module.

Speech is an effective way for communicating and exchanging complex information between humans. Speech signal has involved a great attention in human-computer interaction. Therefore, emotion recognition from speech has become a hot research topic in the field of interacting machines with humans. In this paper, we proposed a novel speech emotion recognition system by adopting multivariate time series handcrafted feature representation from speech signals. Bidirectional echo state network with two parallel reservoir layers has been applied to capture additional independent information. The parallel reservoirs produce multiple representations for each direction from the bidirectional data with two stages of concatenation. The sparse random projection approach has been adopted to reduce the high-dimensional sparse output for each direction separately from both reservoirs. Random over-sampling and random under-sampling methods are used to overcome the imbalanced nature of the used speech emotion datasets. The performance of the proposed parallel ESN model is evaluated from the speaker-independent experiments on EMO-DB, SAVEE, RAVDESS, and FAU Aibo datasets. The results show that the proposed SER model is superior to the single reservoir and the state-of-the-art studies.

Lifelong learning is a long-standing aim for artificial agents that act in dynamic environments, in which an agent needs to accumulate knowledge incrementally without forgetting previously learned representations.We investigate methods for learning from data produced by event cameras and compare techniques to mitigate forgetting while learning incrementally.We propose a model that is composed of both, feature extraction and continuous learning.Furthermore, we introduce a habituationbased method to mitigate forgetting.Our experimental results show that the combination of different techniques can help to avoid catastrophic forgetting while learning incrementally from the features provided by the extraction module.

Clustering based on Adaptive Resonance Theory (ART) has been actively studied. In previous studies, ART-based clustering algorithms with a topological structure have been proposed and showed their superior self-organizing ability. However, this method deteriorates the clustering performance at high noise ratios. In this paper, we propose a multilayer clustering algorithm based on a topological ART-based clustering for improving a noise reduction ability. Simulation experiments show that the proposed algorithm achieves excellent clustering performance on a 2D synthetic dataset in high noise environments.

Falls are a major health issue, particularly among the elderly. Increasing fall events require high service quality and dedicated medical treatment which is an economic burden. In the lack of appropriate care and support, serious injuries caused by fall will cost lives. Therefore, tracking systems with fall detection capabilities are required. Static-view sensors with machine learning techniques for human fall detection have been widely studied and achieved significant results. However, these systems unable to monitor a person if he or she is out of viewing angle which greatly impedes its performance. Mobile robots are an alternative for keeping the person in sight. However, existing mobile robots are unable to operate for a long time due to battery issues and movement constraints in complex environments. In this paper, we proposed a lightweight deep learning vision-based model for human fall detection with an assistive robot to provide assistance when a fall happens. The proposed detection system requires less computational power which can be implemented in a low-cost 2D camera and GPU board for real-time monitoring. The assistive robot equipped with various sensors that can perform SLAM, obstacle avoidance and navigation autonomously. Our proposed system integrates these two sub-systems to compensate for the weakness of each other to constitute a system that robust, adaptable, and high performance. The proposed method has been validated through a series of experiments.

Catastrophic forgetting is a major problem that affects neural networks during progressive learning. In it, the previously learned representation vanishes as the network learns new information. The extreme learning machine is one of the variants of the neural network. It is used in many domains due to fast training and good generalization ability. However, like other neural networks, it suffers from catastrophic forgetting and negative forward and backward transfer during the progression of neurons in incremental learning. The study hypothesizes that it is due to overlapping in hidden neurons and output weights. The global representation by an activation function further supports this hypothesis. To address this, the study proposes a neuron clustering approach to mitigate it in an adaptive class incremental extreme learning machine. The neuron clustering method activates k nearest neurons during learning and testing. It helps to partition the network to select overlapping subnetwork. Experimental results on four food datasets show that the proposed approach reduces negative forward and backward transfer when neurons are added incrementally during progressive learning.

Biological information inspires the advancement of a navigational mechanism for autonomous robots to help people explore and map real-world environments. However, the robot's ability to constantly acquire environmental information in real-world, dynamic environments has remained a challenge for many years. In this paper, we propose a self-organizing adaptive recurrent incremental network that models human episodic memory to learn spatiotemporal representations from novel sensory data. The proposed method termed as SOARIN consists of two main learning process that is active learning and episodic memory playback. For active learning (robot exploration), SOARIN quickly learns and adapts incoming novel sensory data as episodic neurons via competitive Hebbian Learning. Episodic neurons are connecting with each other and gradually forms a spatial map that can be used for robot localization. Episodic memory playback is triggered whenever the robot is in an inactive mode (charging or hibernating). During playback, SOARIN gradually integrates knowledge and experience into more consolidate spatial map structures that can overcome the catastrophic forgetting. The proposed method is analyzed and evaluated in term of map learning and localization through a series of real robot experiments in real-world indoor environments.

Grow-when-required networks such as the Growing Dual-Memory (GDM) networks possess a dynamic network structure, expanding to accommodate new neurons in response to learning novel concepts. Over time, it may be necessary to prune obsolete neurons and/or neural connections to meet performance or resource limitations. GDM networks utilize an age-based pruning strategy, whereby older neurons and neural connections that have not been activated recently are removed. Catastrophic forgetting occurs when knowledge learned by the networks in previous learning iterations is lost due to being overwritten by newer learning iterations, or to the pruning process. In this work, we investigate catastrophic forgetting in GDM networks in response to different pruning strategies. The age-based pruning method was shown to significantly sparsify the GDM network topology while improving the networks ability to recall newly acquired concepts with a slight decrease in performance with respect to older knowledge. A significance-based pruning method was tested as a replacement for the age-based pruning, but was not as effective at pruning even though it performed better at recalling older knowledge.

This paper attempts to solve the typical problems of self-organizing growing network models, i.e. (a) an influence of the order of input data on the self-organizing ability, (b) an instability to high-dimensional data and an excessive sensitivity to noise, and (c) an expensive computational cost by integrating Kernel Bayes Rule (KBR) and Correntropy-Induced Metric (CIM) into Adaptive Resonance Theory (ART) framework. KBR performs a covariance-free Bayesian computation which is able to maintain a fast and stable computation. CIM is a generalized similarity measurement which can maintain a high-noise reduction ability even in a high-dimensional space. In addition, a Growing Neural Gas (GNG)-based topology construction process is integrated into the ART framework to enhance its self-organizing ability. The simulation experiments with synthetic and real-world datasets show that the proposed model has an outstanding stable self-organizing ability for various test environments.

Imitation learning through self-exploration is essential in developing sensorimotor skills. Most developmental theories emphasize that social interactions, especially understanding of observed actions, could be first achieved through imitation, yet the discussion on the origin of primitive imitative abilities is often neglected, referring instead to the possibility of its innateness. This paper presents a developmental model of imitation learning based on the hypothesis that humanoid robot acquires imitative abilities as induced by sensorimotor associative learning through self-exploration. In designing such learning system, several key issues will be addressed: automatic segmentation of the observed actions into motion primitives using raw images acquired from the camera without requiring any kinematic model; incremental learning of spatio-temporal motion sequences to dynamically generates a topological structure in a self-stabilizing manner; organization of the learned data for easy and efficient retrieval using a dynamic associative memory; and utilizing segmented motion primitives to generate complex behavior by the combining these motion primitives. In our experiment, the self-posture is acquired through observing the image of its own body posture while performing the action in front of a mirror through body babbling. The complete architecture was evaluated by simulation and real robot experiments performed on DARwIn-OP humanoid robot.

Quantum-inspired computing is an emerging research area, which has significantly improved the capabilities of conventional algorithms. In general, quantum-inspired hopfield associative memory (QHAM) has demonstrated quantum information processing in neural structures. This has resulted in an exponential increase in storage capacity while explaining the extensive memory, and it has the potential to illustrate the dynamics of neurons in the human brain when viewed from quantum mechanics perspective although the application of QHAM is limited as an autoassociation. We introduce a quantum-inspired multidirectional associative memory (QMAM) with a one-shot learning model, and QMAM with a self-convergent iterative learning model (IQMAM) based on QHAM in this paper. The self-convergent iterative learning enables the network to progressively develop a resonance state, from inputs to outputs. The simulation experiments demonstrate the advantages of QMAM and IQMAM, especially the stability to recall reliability.

The application of multiagent system (MAS) is becoming increasing popular as it allows agents in a system to pool resources together to achieve a common objective. A vital part of the MAS is the teamwork cooperation through the sharing of information and resources among the agents to optimize their efforts in accomplishing given objectives. A critical part of the teamwork effort is the ability to trust each other when executing any task to ensure efficient and successful cooperation. This paper presents the development of a trust estimation model that could empirically evaluate the trust of an agent in MAS. The proposed model is developed using temporal difference learning by incorporating the concept of Markov games and heuristics to estimate trust. Simulation experiments are conducted to test and evaluate the performance of the developed model against some of the recently reported model in the literature. The simulation experiments indicate that the developed model performs better in terms of accuracy and efficiency in estimating trust.

A medical case study related to implantable rotary blood pumps is examined. Five classifiers and two ensemble classifiers are applied to process the signals collected from the pumps for the identification of the aortic valve nonopening pump state. In addition to the noise-free datasets, up to 40% class noise has been added to the signals to evaluate the classification performance when mislabeling is present in the classifier training set. In order to ensure a reliable diagnostic model for the identification of the pump states, classifications performed with and without class noise are evaluated. The multilayer perceptron emerged as the best performing classifier for pump state detection due to its high accuracy as well as robustness against class noise.

An accurate and noninvasive stress assessment from human physiology is a strenuous task. In this paper, a pattern recognition system to learn complex correlates between heart rate variability (HRV) features and salivary stress biomarkers is proposed. Using the Trier social stress test, heart rate and salivary measurements were obtained from volunteers under varying levels of stress induction. Measurements of salivary alpha-amylase and cortisol were used as objective measures of stress, and were correlated with the HRV features using fuzzy ARTMAP (FAM). In improving the predictive ability of the ARTMAPs, techniques, such as genetic algorithms for parameter optimization and voting ensembles, were employed. The ensemble of FAMs can be used for predicting stress responses of salivary alpha-amylase or cortisol using heart rate measurements as the input. Using alpha-amylase as the stress indicator, the ensemble was able to classify stress from heart rate features with 75% accuracy, and 80% accuracy when cortisol was used.

A hybrid intelligent model comprising a modified fuzzy min-max (FMM) clustering neural network and a modified clustering tree (CT) is developed. A review of clustering models with rule extraction capabilities is presented. The hybrid FMM-CT model is explained. We first use several benchmark problems to illustrate the cluster evolution patterns from the proposed modifications in FMM. Then, we employ a case study with real data related to power quality monitoring to assess the usefulness of FMM-CT. The results are compared with those from other clustering models. More importantly, we extract explanatory rules from FMM-CT to justify its predictions. The empirical findings indicate the usefulness of the proposed model in tackling data clustering and power quality monitoring problems under different environments.

Human-like-motion performed by robots can have a contribution to exert a strong influence on human-robot interaction, because bodily expressions convey important and effective information. If the robots could adapt the features of human behavior to their motions and skills, the communication would become more smooth and natural. In this paper, we develop a posture measurement system for a robot imitation using a 3D image sensor. This paper proposes a method of robot posture generation based on a steady-state genetic algorithm (SSGA). SSGA is one of evolutionary optimization methods using selection, mutation, and crossover operators. Since SSGA is a simplified model, it is easy to implement into a real-time processing. Furthermore, we apply a continuous model of generation for an adaptive search in dynamical environment.

In this study, we aim to develop a system for improving daily lives of elderly people to ensure health. In order to realize an enriched life style among elderly people, daily health care is important. Therefore, we have proposed a system where robot partners will assist in exercising activity among elderly people. This paper proposes a method of relational trust modeling based on reinforcement learning. We apply a concept of relational trust defined as the expectation that a person is disposed to act in a trustworthy manner toward "me," no matter what the person does to others. In the experiment, we discuss the effectiveness of relational trust for robot communication.

Complex-valued neuron is one of the significant and effective innovation in artificial neural networks. It is able to deal with multi-valued pattern and oscillator models. With regard as associative memory, several types of complex-valued artificial neural associative memories have developed, and confirmed its superior abilities. Conventionally, we have developed Quantum-Inspired Mulitidirectional Associative Memory (QMAM). This model demonstrates quantum information processing in neural structures results in an exponential increase in storage capacity and can explain the extensive memory and inferencing capabilities of humans. This model is applied a fuzzy inference to weight matrix to satisfy parallelism and unitarity. In this paper, we introduce Quantum-Inspired Complex-Valued Multidirectional Associative Memory (QCMAM) to handle multi-valued information. In addition, the mathematical proofs of parallelism and unitarity for a complex-valued model are presented. The simulation experiments show that QCMAM has superior abilities comparing with conventional model.

Emotion and personality are significant factors in communication. In general, during a decision making process in human-human communication, emotional factors will be affected not only the logical thinking, but also the emotional responses. Furthermore, personality gives individual differences among people in behavior patterns, cognitive process and emotional responses. In this paper, we propose the three stages (core affects, emotion and mood) robotic emotional model with OCEAN model as the personality factors based on 2D (Pleasant-Arousal) scaling model. The emotion states in proposed model are represented on pleasant-arousal plane. The results from simulation experiment show that the proposed model is able to generate the different emotional properties based on the personality factors.

In order to keep healthy health of elderly people, daily exercise is an important factor. Therefore, we have developed an exercise support system with robot partner to provide the daily exercise program. Furthermore, Human-like-motion can have a contribution to exert a strong influence on the person through the human-robot interaction, because bodily expressions convey important and effective information. If robots could adapt the features of human behavior to their motions and skills, the communication would become more smooth and natural. In this paper, we propose a learning structure for imitation learning.

Quantum-Inspired Computational Intelligence based on quantum postulates is an emerging research area that exploit the parallelism of quantum mechanics. However, existing research efforts are limited to theoretical simulations and has not been implemented in robot application. With regards as Human-Robot Interaction, associative memory become essential for mutual communication. However, associative memory always suffers from limited memory capacity and high sensitivity to noise and the ability to recall from multi-modal sensory inputs. In this paper, we propose a Quantum-Inspired Multidirectional Associative Memory. This is the first attempt to overcome these two problems effectively with robot application.

Recent works involved in enhancing the learning convergence of reinforcement learning (RL) in mobile robot navigation have investigated methods to obtain knowledge from efficiently exploring the robot's environment. In RL, this knowledge is highly desirable to reduce the high number of interactions required for updating the value function and to eventually find an optimal or suboptimal policy for the agent. In this work, we propose a curiosity-based topological RL (CBT-RL) algorithm that makes use of the topological relationships among the observed states of the environment in which the agent acts. This algorithm builds an incremental topological map of the environment using Instantaneous Topological Map (ITM) model, which we use for facilitating value function updates as well as providing a guided exploration. We evaluate our algorithm against the original Q-Learning and Influence Zone algorithms in static and dynamic environments.

In this paper, the application of a hybrid model combining the fuzzy min-max (FMM) neural network and the classification and regression tree (CART) to human activity recognition is presented. The hybrid FMM-CART model capitalizes the merits of both FMM and CART in data classification and rule extraction. To evaluate the effectiveness of FMM-CART, two data sets related to human activity recognition problems are conducted. The results obtained are higher than those reported in the literature. More importantly, practical rules in the form of a decision tree are extracted to provide explanation and justification for the predictions from FMM-CART. This outcome positively indicates the potential of FMM-CART in undertaking human activity recognition tasks.

In this paper we present an approach for on-line and incremental learning of human motion patterns through continuous observation of motion using novel Topological Gaussian Adaptive Resonance Hidden Markov Model (TGART-HMM). The observed human motion patterns are encoded in a novel modified version of Hidden Markov Model (HMM) called TGART-HMM. The on-line learning process consists of updating the structure of Hidden Markov Model using a topology-learning mechanism based on Gaussian Adaptive Resonance Theory (GART). The model size is adaptable based on the observed motion patterns. The resulting HMM structure is a graph where each node represents an encoded motion pattern. The parameters of TGART-HMM are updated incrementally to incorporate incessant motion patterns. The algorithm is tested on motion captured data to test the efficacy of the system.

The proposed Fast Incremental Slow Feature Analysis (F-IncSFA) which is considered as unsupervised learning and it can be used for extracting the features. The featurescan represent the fundamental components of the modifications in different aspect and especially in posing and temporally firms and consistent even in high-dimensional input like signal, video, etc. Here, we addressed a development in SFA algorithm as compare with latest one [17] by combining Candid Covariance-Free Incremental Principle components Analysis (CCIPCA) and Minor Components Analysis (MCA).The proposed F-IncSFA can adapts along with non-stationary environments and unlike the latest SFA, which has two times using CCIPCA, has one time using CCIPCA in its algorithm which makes the method simpler yet efficient. We examine the proposed approach by using some video sequences of humanoid robot and also it is compared with CCIPCA in several experiments and the result indicates that it indeed has superior outcome and impart informative slow features that is representing significant abstract from possessions of non-stationary environment and poses. We successfully apply the F-IncSFA on the high-dimensional video and extract abstract object data. We extend our F-IncSFA to networks in hierarchical model, and apply it for extraction of features in the information obtained from high-dimensional video and the results were promising.

A small change in image will cause a dramatic change in signals. Visual system is required to be able to ignore these changes, yet specific enough to perform recognition. This work intends to provide biological-backed insights into 2D translation and scaling invariance and 3D pose-invariance without imposing strain on memory and with biological justification. The model can be divided into lower and higher visual stages. Lower visual stage models the visual pathway from retina to the striate cortex (V1), whereas the modeling of higher visual stage is mainly based on current psychophysical evidences.

The QMS (Quantum Morphogenetic System) is a mathematical model devised to give a different more elegant and intuitive perspective of the pre-processing by assuming a non-Euclidean geometry. Input image is projected to the feature vector in terms of its contra-variant components which are interdependent. Image reconstruction is achieved by parallelogram summation of the vector of those components. View-Invariant recognition is achieved through the manipulation of association matrix which in geometrical perspective is a deformation of space towards a canonical point for all vectors which are being considered the same.

Hopfield network requires that state vectors of images to be orthogonal to eliminate cross-talk. In practical sense, it cannot be achieved. A solution is to orthogonalize the state vectors with each other as a pre-processing. The QMS (Quantum Morphogenetic System) is a mathematical model devised to give a different more elegant and intuitive perspective of the pre-processing by assuming a non-Euclidean geometry. Input image is projected to the feature vector in terms of its contra-variant components which are interdependent. Image reconstruction is achieved by parallelogram summation of the vector of those components. Besides, an extension to Holonomic brain model and tensor network theory is discussed.

In this paper, an accurate and effective probabilistic plurality voting method to combine outputs from multiple simplified fuzzy ARTMAP (SFAM) classifiers is presented. Five ELENA benchmark problems and five medical benchmark data sets have been used to evaluate the applicability and performance of the proposed probabilistic ensemble simplified fuzzy ARTMAP (PESFAM) network. Among the five benchmark problems in ELENA project, PESFAM outperforms the SFAM and multi-layer perceptron (MLP) classifier. In addition, the effectiveness of the proposed PESFAM is delineated in medical diagnosis applications. For the medical diagnosis and classification problems, PESFAM achieves 100 percent in accuracy, specificity, and sensitivity based on the 10-fold crossvalidation and these results are superior to those from other classification algorithms. In addition, a posteri probability of the predicted class can be used to measure the prediction reliability of PESFAM. The experiments demonstrate the potential of the proposed multiple SFAM classifiers in offering an optimal solution to the data-ordering problem of SFAM implementation and also as an intelligent medical diagnosis tool.

This paper presents two novel approaches to determine optimum growing multi-experts network (GMN) structure. The first method called direct method deals with expertise domain and levels in connection with local experts. The growing neural gas (GNG) algorithm is used to cluster the local experts. The concept of error distribution is used to apportion error among the local experts. After reaching the specified size of the network, redundant experts removal algorithm is invoked to prune the size of the network based on the ranking of the experts. However, GMN is not ergonomic due to too many network control parameters. Therefore, a self-regulating GMN (SGMN) algorithm is proposed. SGMN adopts self-adaptive learning rates for gradient-descent learning rules. In addition, SGMN adopts a more rigorous clustering method called fully self-organized simplified adaptive resonance theory in a modified form. Experimental results show SGMN obtains comparative or even better performance than GMN in four benchmark examples, with reduced sensitivity to learning parameters setting. Moreover, both GMN and SGMN outperform the other neural networks and statistical models. The efficacy of SGMN is further justified in three industrial applications and a control problem. It provides consistent results besides holding out a profound potential and promise for building a novel type of nonlinear model consisting of several local linear models.

We present some experiments with a new class of variations of mutation to accelerate the convergence of PSO. These robust mutation variations are tested on benchmark problems and the results show a significant improvement as compared to the original particle swarm optimization algorithm.

This paper presents the design of ant-inspired control strategies that mimic the ant colony foraging behavior. Inter-agent communication, in particular explicit communication is applied to create multiple cooperating mobile robots in order to accomplish the foraging task. Explicit communication can significantly multiply the capabilities and effectiveness of teams of robotic systems. However, the drawback is it introduces more interference among robots at the goal and home region in comparison to robotic system without the implementation of inter-agent communication. Thus, to investigate how well the explicit communication can be adopted in designing efficient cooperative mobile robotic system, experiments were carried out on the simulated robots. The efficiency of the strategies are measured in terms of three criteria: time, density of robots and interference.

In this paper, the application of PID control to three non-minimum phase plants has been examined through a series of simulation studies. Initially, Ziegler-Nichols rules have been used for tuning the PID controller. Later, an intelligent strategy - Differential Evolution has been proposed in order to tune automatically the PID controller. The effectiveness of the proposed method is investigated and it is found that the systems are with better results, in the aspects of rise time, settling time, steady-state error, peak overshoot and maximum undershoot.