Search

This paper presents a range of novel reinforcement learning algorithms derived from the actor–critic approach. These modified algorithms effectively utilize the available information to enhance performance. Our proposed framework introduces several key components to the traditional actor–critic model, including an underlying model learner, selector, tuner, and estimator. The estimator employs an approximate value function and the learned underlying model to estimate the values of all actions at the next state. The selector approximates the optimal action at the next state, which is then utilized by the actor to optimize its policy. In contrast to the conventional actor–critic algorithm where the actor focuses solely on policy optimization and the critic performs value-function approximation and policy evaluation, our selector–actor–critic algorithm employs a selector to approximate the optimal action at the current state, thereby influencing the actor's policy updates. Furthermore, our tuner–actor–critic algorithm incorporates a critic and a model-learner to approximate the action-value function and the dynamics of the underlying environment, respectively. The tuner then utilizes this information to adjust the value of the current state–action pair. In the estimator–selector–actor–critic algorithm, we develop intelligent agents based on the concepts of lookahead and intuition. Lookahead is utilized in estimating the values of available actions at the next state, while intuition guides the maximization of the probability of selecting the approximate optimal action. Through simulation experiments, we evaluate the performance of these algorithms, and the results demonstrate the superiority of the estimator–selector–actor–critic approach over other existing algorithms. [ABSTRACT FROM AUTHOR]

Electric vehicles (EVs) are gaining popularity as an efficient and environment-friendly transportation alternative. However, as the number of EVs on the road increases, the demand for electric power to charge them also rises, escalating challenges for the power industry. This requires accurate profiling of EV charging patterns to make informed decisions in future integrated energy systems. To address this problem, various traditional algorithms have been used to forecast the electricity demand for EV charging in the future. Recently, AI techniques, such as neural networks, machine learning, and deep learning, have shown promise in leveraging extensive historical datasets to identify patterns and generate precise predictions. This research paper introduces a novel implementation of an XGBoost regression tree-based algorithm applied to three real-world datasets to estimate the future electric load of EVs. Experimental results demonstrate the superior performance of the proposed algorithm compared to eight state-of-the-art machine learning algorithms, evaluated using the root-mean-squared error and mean absolute error. Furthermore, the proposed algorithm outperforms previous studies in this field, highlighting its effectiveness in forecasting EV charging power requirements. [ABSTRACT FROM AUTHOR]

In today's constantly advancing economy, fast and efficient load frequency control (LFC) schemes are imperative for stable power systems operation, whether conventional or new. Despite recent advances in this domain, professional engineers, to this day, face significant obstacles in dealing with parametric uncertainty and the rejection of external disturbances. To ensure high-quality, resilient, and consistent electrical power, the developed controllers must operate efficiently, i.e., counteract frequency and tie-line power deviations. In light of this, the current study presents a novel strategy employing several machine learning algorithms to remedy the LFC dilemma, estimating the optimal controller gain values in a real-time environment. A comprehensive case study incorporating random load perturbations exemplified the trained controllers' accuracy, robustness, and adaptability. The proposed online controllers, integrated into emerging power systems with several adjoining regions, renewable sources, and nonlinearities, outperform benchmark procedures. A rank-based statistical analysis implementing the Friedman test reveals Random Forest to be the most effective of the studied algorithms in terms of providing the lowest integral-criteria error values while preserving the minimal peak overshoot, undershoot, and settling duration. This paper demonstrates the significant potential of the suggested framework in addressing the LFC predicament in relevant sectors. [ABSTRACT FROM AUTHOR]

Healthcare sensors represent a valid and non-invasive instrument to capture and analyse physiological data. Several vital signals, such as voice signals, can be acquired anytime and anywhere, achieved with the least possible discomfort to the patient thanks to the development of increasingly advanced devices. The integration of sensors with artificial intelligence techniques contributes to the realization of faster and easier solutions aimed at improving early diagnosis, personalized treatment, remote patient monitoring and better decision making, all tasks vital in a critical situation such as that of the COVID-19 pandemic. This paper presents a study about the possibility to support the early and non-invasive detection of COVID-19 through the analysis of voice signals by means of the main machine learning algorithms. If demonstrated, this detection capacity could be embedded in a powerful mobile screening application. To perform this important study, the Coswara dataset is considered. The aim of this investigation is not only to evaluate which machine learning technique best distinguishes a healthy voice from a pathological one, but also to identify which vowel sound is most seriously affected by COVID-19 and is, therefore, most reliable in detecting the pathology. The results show that Random Forest is the technique that classifies most accurately healthy and pathological voices. Moreover, the evaluation of the vowel /e/ allows the detection of the effects of COVID-19 on voice quality with a better accuracy than the other vowels. [ABSTRACT FROM AUTHOR]

In this research, evolutionary formula-based models were developed for designing of eco-friendly concrete containing recycled polyethylene terephthalate (PET). To do so, evolutionary artificial intelligence (AI) approach was implemented based on the integration of the multivariate adaptive regression splines (MARS) and particle swarm optimization algorithm for estimation of compressive strength (CS) of eco-friendly concrete containing recycled PET. The experimental database consisting 320 records comprising mixture components at different ages was collected from published documents. The capability and efficiency of proposed model were validated through different AI methods including linear regression, extreme learning machine, random forest, M5p model tree and standalone MARS. Performance metrics indicated that proposed evolutionary model with the optimized parameters outperformed other benchmark models in term of accuracy. Uncertainty analysis of the standalone and hybridized models were also applied using Monte-Carlo simulation to prove that the model has less uncertainty in the prediction of the CS compared to those benchmark models. The findings of the paper presented the superiority of the model's development in constructing reasonable and robustness evolutionary model for formulation of CS of eco-friendly concrete. [ABSTRACT FROM AUTHOR]

Chinese Sign Language (CSL) offers the main means of communication for the hearing impaired in China. Sign Language Recognition (SLR) can shorten the distance between the hearing-impaired and healthy people and help them integrate into the society. Therefore, SLR has become the focus of sign language application research. Over the years, the continuous development of new technologies provides a source and motivation for SLR. This paper aims to cover the most recent approaches in Chinese Sign Language Recognition (CSLR). With a thorough review of superior methods from 2000 to 2019 in CSLR researches, various techniques and algorithms such as scale-invariant feature transform, histogram of oriented gradients, wavelet entropy, Hu moment invariant, Fourier descriptor, gray-level co-occurrence matrix, dynamic time warping, principal component analysis, autoencoder, hidden Markov model (HMM), support vector machine (SVM), random forest, skin color modeling method, k-NN, artificial neural network, convolutional neural network (CNN), and transfer learning are discussed in detail, which are based on several major stages, that is, data acquisition, preprocessing, feature extraction, and classification. CSLR was summarized from some aspect as follows: methods of classification and feature extraction, accuracy/performance evaluation, and sample size/datasets. The advantages and limitations of different CSLR approaches were compared. It was found that data acquisition is mainly through Kinect and camera, and the feature extraction focuses on hand's shape and spatiotemporal factors, but ignoring facial expressions. HMM and SVM are used most in the classification. CNN is becoming more and more popular, and a deep neural network-based recognition approach will be the future trend. However, due to the complexity of the contemporary Chinese language, CSLR generally has a lower accuracy than other SLR. It is necessary to establish an appropriate dataset to conduct comparable experiments. The issue of decreasing accuracy as the dataset increases needs to resolve. Overall, our study is hoped to give a comprehensive presentation for those people who are interested in CSLR and SLR and to further contribute to the future research. [ABSTRACT FROM AUTHOR]

In this paper, we design and implement a human-machine interaction application, which enables a visually challenged person to locate and manipulate personal objects in her/his neighborhood. In this setting, we need to develop a tool (embedded in a mobile phone) which is capable of sensing, computing, and guiding the human arm toward the object. This involves solving the following two subproblems: (1) recognition of objects in the input images, and (2) generating control signals, to guide the human for navigation, to reach the desired destination. For the former subproblem, we adapt the bag-of-words framework for recognition and matching on mobile phones. For the latter subproblem, we have developed a moment-based human servoing algorithm which is able to generate commands that help the visually impaired human to localize his hand with respect to the object of interest. All necessary computations take place on the mobile phone. The proposed object recognition and vision-based control design are deployed on a low-/mid-end mobile phone. This can lead to a wide range of applications. With our proposed design and implementation, we demonstrate that our application is effective and accurate, with a high reliability of convergence for different experimental settings. [ABSTRACT FROM AUTHOR]

One impeding factor for determining the correct number of solar filaments in H- $$\upalpha $$ solar images is its broken appearance, which, in turn, affects the accuracy of tracking it and extracting its attributes accurately. By integrating image processing (IP) and artificial intelligence techniques, these fragmented threads have been more effectively merged. In this paper, different IP methods are utilized to extract the filament attributes, where these attributes act as inputs to a neural network to obtain a 92% true-positive rate. The method introduced in this work is fully automated and not threshold-based-unlike in previous works. [ABSTRACT FROM AUTHOR]