Artificial Intelligence Paradigm Research Articles

A cutting-edge artificial intelligence paradigm for entertainment-infused music recommendations

Personalized and fascinating music recommendations are becoming increasingly in demand as the advent of technology continues to change the way that people consume music. Traditional music recommendation systems primarily rely on user listening history and preferences, often neglecting the emotional and experiential dimensions that make music a deeply personal and entertaining endeavour. This research introduces a music recommendation system after classifying the tracks using a novel Gravitational Search Optimized Recursive Neural Networks (GS-RNN) approach. GS-RNN addressed this gap by integrating the Gravitational Search Algorithm (GSA) with Recursive Neural Networks (RNN) to create a content-based recommendation system that assesses auudio signal similarity. Evaluation metrics, including accuracy (80%), logarithmic loss (0.85), precision (84%), recall (83%), and F1-score (88%), demonstrate GS-RNN’s superiority over existing techniques. Genre-specific accuracy analysis further underscores the model’s capability to suggest songs within the same genre. Overall, GS-RNN presents a novel and effective paradigm for music recommendations.

Reinforcement Learning Algorithms and Applications in Healthcare and Robotics: A Comprehensive and Systematic Review.

Reinforcement learning (RL) has emerged as a dynamic and transformative paradigm in artificial intelligence, offering the promise of intelligent decision-making in complex and dynamic environments. This unique feature enables RL to address sequential decision-making problems with simultaneous sampling, evaluation, and feedback. As a result, RL techniques have become suitable candidates for developing powerful solutions in various domains. In this study, we present a comprehensive and systematic review of RL algorithms and applications. This review commences with an exploration of the foundations of RL and proceeds to examine each algorithm in detail, concluding with a comparative analysis of RL algorithms based on several criteria. This review then extends to two key applications of RL: robotics and healthcare. In robotics manipulation, RL enhances precision and adaptability in tasks such as object grasping and autonomous learning. In healthcare, this review turns its focus to the realm of cell growth problems, clarifying how RL has provided a data-driven approach for optimizing the growth of cell cultures and the development of therapeutic solutions. This review offers a comprehensive overview, shedding light on the evolving landscape of RL and its potential in two diverse yet interconnected fields.

Examining Ethical Aspects of AI: Addressing Bias and Equity in the Discipline

he rapid progress in implementing Artificial Intelligence (AI) across various domains such as healthcare decision-making, medical diagnosis, and others has raised significant concerns regarding the fairness and bias embedded within AI systems. This is particularly crucial in sectors like healthcare, employment, criminal justice, credit scoring, and the emerging field of generative AI models (GenAI) producing synthetic media. Such systems can lead to unfair outcomes and perpetuate existing inequalities, including biases ingrained in the synthetic data representation of individuals.This survey paper provides a concise yet comprehensive examination of fairness and bias in AI, encompassing their origins, ramifications, and potential mitigation strategies. We scrutinize sources of bias, including data, algorithmic, and human decision biases, shedding light on the emergent issue of generative AI bias where models may replicate and amplify societal stereotypes. Assessing the societal impact of biased AI systems, we spotlight the perpetuation of inequalities and the reinforcement of harmful stereotypes, especially as generative AI gains traction in shaping public perception through generated content.Various proposed mitigation strategies are explored, with an emphasis on the ethical considerations surrounding their implementation. We stress the necessity of interdisciplinary collaboration to ensure the effectiveness of these strategies. Through a systematic literature review spanning multiple academic disciplines, we define AI bias and its various types, delving into the nuances of generative AI bias. We discuss the adverse effects of AI bias on individuals and society, providing an overview of current approaches to mitigate bias, including data preprocessing, model selection, and post-processing. Unique challenges posed by generative AI models are highlighted, underscoring the importance of tailored strategies to address them effectively.Addressing bias in AI necessitates a holistic approach, involving diverse and representative datasets, enhanced transparency, and accountability in AI systems, and exploration of alternative AI paradigms prioritizing fairness and ethical considerations. This survey contributes to the ongoing discourse on developing fair and unbiased AI systems by outlining the sources, impacts, and mitigation strategies related to AI bias, with a particular focus on the burgeoning field of generative AI.

The Evolving Regulatory Paradigm of AI in MedTech: A Review of Perspectives and Where We Are Today.

Artificial intelligence (AI)-enabled technologies in the MedTech sector hold the promise to transform healthcare delivery by improving access, quality, and outcomes. As the regulatory contours of these technologies are being defined, there is a notable lack of literature on the key stakeholders such as the organizations and interest groups that have a significant input in shaping the regulatory framework. This article explores the perspectives and contributions of these stakeholders in shaping the regulatory paradigm of AI-enabled medical technologies. The formation of an AI regulatory framework requires the convergence of ethical, regulatory, technical, societal, and practical considerations. These multiple perspectives contribute to the various dimensions of an evolving regulatory paradigm. From the global governance guidelines set by the World Health Organization (WHO) to national regulations, the article sheds light not just on these multiple perspectives but also on their interconnectedness in shaping the regulatory landscape of AI.

The evolution of artificial intelligence: problems and prospects of rational cognition

Objective: This article undertakes a comprehensive exploration of the constructivist paradigm in artificial intelligence (AI) development, aiming to uncover how constructivist perspectives shape our understanding of AI. It delves into the evolution of AI thought, emphasizing the significance of constructivist epistemology in comprehending AI's philosophical and cognitive dimensions.
 Method: The study employs a variety of philosophical methodologies, including historical-philosophical analysis, comparative analysis of philosophical teachings, and a system-structural dialectical approach. These methods facilitate an in-depth examination of AI's conceptual intricacies within a constructivist framework, focusing on the relationship between artificial and natural intelligence and the epistemological implications of AI.
 Results: The investigation reveals that the main challenge in AI research is the absence of clear problem-solving rules, highlighting the current limitations of human self-knowledge in logical and emotional intelligence. It showcases AI's vast capabilities, from extensive knowledge bases to real-time processing, and emphasizes AI's role in enhancing human cognitive processes.
 Conclusions: Artificial intelligence, as a construct of human intellect, mirrors the capacity for design and creativity inherent in human thought. The study underscores AI's foundational role in the epistemology of science and technology, advocating for a holistic understanding of the human brain as a dynamic system to further our grasp of AI and its cognitive potential.

The connectivity degree controls the difficulty in reservoir design of random boolean networks.

Reservoir Computing (RC) is a paradigm in artificial intelligence where a recurrent neural network (RNN) is used to process temporal data, leveraging the inherent dynamical properties of the reservoir to perform complex computations. In the realm of RC, the excitatory-inhibitory balance b has been shown to be pivotal for driving the dynamics and performance of Echo State Networks (ESN) and, more recently, Random Boolean Network (RBN). However, the relationship between b and other parameters of the network is still poorly understood. This article explores how the interplay of the balance b, the connectivity degree K (i.e., the number of synapses per neuron) and the size of the network (i.e., the number of neurons N) influences the dynamics and performance (memory and prediction) of an RBN reservoir. Our findings reveal that K and b are strongly tied in optimal reservoirs. Reservoirs with high K have two optimal balances, one for globally inhibitory networks (b < 0), and the other one for excitatory networks (b > 0). Both show asymmetric performances about a zero balance. In contrast, for moderate K, the optimal value being K = 4, best reservoirs are obtained when excitation and inhibition almost, but not exactly, balance each other. For almost all K, the influence of the size is such that increasing N leads to better performance, even with very large values of N. Our investigation provides clear directions to generate optimal reservoirs or reservoirs with constraints on size or connectivity.

Design of Planar Multilayer Devices for Optical Filtering Using Surrogate Model Based on Artificial Neural Network

Artificial intelligence paradigms hold significant potential to advance nanophotonics. This study presents a novel approach to designing a plasmonic absorber using an artificial neural network as a surrogate model in conjunction with a genetic algorithm. The methodology involved numerical simulations of multilayered metal–dielectric plasmonic structures to establish a dataset for training an artificial neural network (ANN). The results demonstrate the proficiency of the trained ANN in predicting reflectance spectra and its ability to generalize intricate relationships between desired performance and geometric configurations, with values of correlation higher than 98% in comparison with ground-truth electromagnetic simulations. Furthermore, the ANN was employed as a surrogate model in a genetic algorithm (GA) loop to achieve target optical behaviors. The proposed methodology provides a powerful means of inverse designing multilayered metal–dielectric devices tailored for visible band wavelength filtering. This research demonstrates that the integration of AI-driven approaches in nanophotonics leads to efficient and effective design strategies.

Cognitive Computing Emulating Human Intelligence in AI Systems

Cognitive computing represents a groundbreaking paradigm in artificial intelligence (AI) systems, aiming to emulate and replicate the intricate processes of human intelligence. This article explores the fundamental principles, methodologies, and applications of cognitive computing, shedding light on how it transforms traditional AI approaches. By drawing inspiration from human cognition, cognitive computing systems leverage advanced algorithms, neural networks, and machine learning techniques to emulate complex cognitive functions such as perception, reasoning, and problem-solving.

Opportunities and challenges for the application of artificial intelligence paradigms into the management of endemic viral infections: The example of Chronic Hepatitis C Virus

AbstractDespite the advent of direct‐acting antiviral agents (DAAs) as a definitive therapy for chronic hepatitis C virus (HCV) infection, the burden of the disease remains globally elevated. The emerging big data on different HCV paradigms fostered the introduction of artificial intelligence/machine learning (AI/ML) applications to help decrease that burden by providing more optimised strategies for early diagnosis and treatment prioritisation. The current review provides descriptive and analytical insight into the recently published AI/ML applications in five medical aspects of HCV infection. In addition, it highlights the opportunities these powerful tools offer in designing national health policies that prioritise HCV patients for the costly DAAs and developing broadly neutralising HCV antibodies. Finally, this paper highlights the challenges encountered in developing and applying these AI/ML models to clinical practice and suggests schemes to overcome some of them. The presented models were primarily evaluated using the Matthews correlation coefficient and the F1‐score to make a more reliable inference about their predictive power under imbalanced datasets. Many published AI/ML applications offered great utilities for predicting novel HCV treatments and prioritising patients for DAAs receipt, especially in settings of limited resources and high HCV burden. Some outperformed the classical diagnostic tools, such as third‐generation serological tests, alpha‐fetoprotein, and ultrasound, in detecting HCV infections and early HCV‐associated hepatocellular carcinoma, respectively. However, further statistical and clinical validation of AI/ML models is highly advocated before incorporating these applications into clinical practice.

Visual Questioner using Bert Transformer

The Visual Questioner, an innovative machine learning endeavor, represents a cutting-edge project designed to process images and facilitate user interaction through AI-generated responses to inquiries. Focused on the fusion of visual recognition and natural language understanding, the overarching objective of this project is to cultivate a robust model capable of discerning intricate details within an image and, subsequently, formulating coherent textual responses to user-generated questions. At the core of the Visual Questioner lies the imperative to bridge the semantic gap between visual content and linguistic expression, enhancing the depth of comprehension and interaction within the artificial intelligence paradigm. By harnessing advanced algorithms and neural network architectures, this initiative aspires to elevate the sophistication of image-based question-answering systems, presenting a pivotal advancement in the realm of human-machine communication. This research not only delves into the technical intricacies of image understanding but also underscores the imperative role of natural language generation, thereby contributing substantively to the evolving landscape of AI applications. As we navigate through the complexities of this project, its potential implications span diverse domains, ranging from human-computer interaction to autonomous systems and beyond, underscoring the multifaceted significance of the Visual Questioner in contemporary artificial intelligence research.

Demystifying AI: Navigating the Balance between Precision and Comprehensibility with Explainable Artificial Intelligence

Integrating Artificial Intelligence (AI) into daily life has brought transformative changes, ranging from personalized recommendations on streaming platforms to advancements in medical diagnostics. However, concerns about the transparency and interpretability of AI models, intense neural networks, have become prominent. This paper explores the emerging paradigm of Explainable Artificial Intelligence (XAI) as a crucial response to address these concerns. Delving into the multifaceted challenges posed by AI complexity, the study emphasizes the critical significance of interpretability. It examines how XAI is fundamentally reshaping the landscape of artificial intelligence, seeking to reconcile precision with the transparency necessary for widespread acceptance.

WTDP-Shapley: Efficient and Effective Incentive Mechanism in Federated Learning for Intelligent Safety Inspection

Artificial Intelligence (AI) has been widely applied for Safety Inspection in a number of industrial domains. However, individual company usually could not provide sufficient data to support well-trained AI models. Federated Learning, as a new AI paradigm, enables a number of participants to contribute training data to co-create high-performance models without compromising data privacy. However, an effective incentive mechanism is essential to encourage participants to contribute high-quality data, and the fundamental of the incentive mechanism is to evaluate participants&#x0027; contribution fairly. Shapley Value (SV) is a well-known approach to evaluate individual&#x0027;s marginal contribution in a coalition, but the canonical SV calculation and its available variants are very costly. In this paper, we proposed an FL framework to enable a number of natural gas companies from different cities to jointly train an object detection computer vision deep learning model for the purpose of identifying potential hazards, without sharing their confidential inspection photos directly. We improve state-of-the-art SV algorithm by proposing Weighted Truncation (WT) for unnecessary computations, and go further with Dynamic Programming (WTDP), achieving better trade-off between efficiency and accuracy. Based on our proposed WTDP-Shapley participant contribution evaluation approach, an effective end-to-end incentive mechanism is designed by leveraging knowledge of both data scientists and domain experts. According to our experiments, it could encourage participants to contribute scarce photos with potential hazards, and thus co-create a high-performance AI model to identify various hazards accurately for residential natural gas installation safety inspection.

Photonic Neural Networks Based on Integrated Silicon Microresonators

Recent progress in artificial intelligence (AI) has boosted the computational possibilities in fields in which standard computers are not able to perform adequately. The AI paradigm is to emulate human intelligence and therefore breaks the familiar architecture on which digital computers are based. In particular, neuromorphic computing, artificial neural networks (ANNs), and deep learning models mimic how the brain computes. There are many applications for large networks of interconnected neurons whose synapses are individually strengthened or weakened during the learning phase. In this respect, photonics is a suitable platform for implementing ANN hardware owing to its speed, low power dissipation, and multi-wavelength opportunities. One photonic device that could serve as an optical neuron is the optical microring resonator. Indeed, microring resonators exhibit a nonlinear response and the capability for optical energy storage, which can be used to implement fading memory. In addition, their characteristic resonant behavior makes them extremely sensitive to input wavelengths, which promotes wavelength division multiplexing (WDM) applications and enables their use as WDM-based synapses (weight banks) in the linear regime. Remarkably, using silicon photonics, photonic integrated circuits can be fabricated in volume and with integrated electronics onboard. For these reasons, here, we describe the physics of silicon microring resonators and arrays of microring resonators for application in neuromorphic computing. We describe different types of ANNs, from feedforward networks to photonic extreme learning machines, and reservoir computing. In addition, we discuss hybrid systems in which silicon microresonators are coupled with other active materials. This review introduces the basics and discusses the most recent developments in the field.

Chatbots: A New Digital Teaching Tool Paradigm in Artificial Intelligence (AI) Technology Among Secondary School Teachers

Teachers around the world are committed to improving student-centered learning, intending to enhance the digital teaching tool in the classroom, in tandem with the rapid evolution of technology. In addition, as a teacher, education is one of the sectors that play an essential role in preparing students for the changes, as students begin to recognize what they are learning on the first day of formal education. Consequently, exploring teachers’ experience and reviewing the challenges in teaching, is in line with the current digital era. The implementation has significance important in the scenario of Industrial Revolution 4.0 (IR4.0). One of the pillars, Artificial Intelligence (AI) Technology, which is defined as the ability of a computer or a computer-controlled robot to perform tasks that are normally performed by humans has given a dynamic viewpoint in various fields, especially in teaching. Nowadays, decision-making is simplified by artificial decision-making, which does not require the presence of a human. As a result, Chatbots as in AI Technology provide a new paradigm for teachers at school, and the mushrooming of digital mediums in education enhances the capability of teaching by using chatbot applications in the lessons. The purpose of this study is to explore Chat Bots in AI Technology as a digital teaching tool among Secondary School Teachers. The researcher used a qualitative case study guide with the purposive sampling method with three research questions to gather data through semi-structured interview questions and document review. Next, the researcher used manual coding to create themes for the findings. All in all, the informants are sharing the concept, the use, and the challenges based on content representation, the students’ interactions in previous experiences, assessments, and feedback. How teachers respond in depth led to the researcher discovering the need for training, professional development, and future classrooms for secondary education.

Artificial Intelligence Techniques in Bioinformatics: Unravelling Complex Biological Systems

This paper reviews the integration of Artificial Intelligence (AI) techniques with bioinformatics, focusing on its applications in absorbing large amounts of biological data and understanding complex biological systems. It covers various AI paradigms, including data mining, machine learning, deep learning, and adaptive algorithms, and their applications in drug discovery, functional genomics, targeted medicine, protein structure prediction, and genomic sequence analysis. The paper emphasizes the role of AI algorithms and biological data in improving knowledge extraction, pattern recognition, and predictive modelling in natural settings. Furthermore, this research evaluates the difficulties and potential applications of AI in bioinformatics, including limitations with data quality, AI model interpretability, integrating multi-omics data, and ethical issues. In summary, this review fabricates the most recent state-of-the-art AI approaches in bioinformatics and offers researchers, practitioners, and stakeholders a road map for utilizing AI developments to effectively decipher biological systems' complexity and produce groundbreaking discoveries and medical applications

Investigation of the solubility of anticancer drugs in the supercritical solvent for development of innovative drug delivery systems; artificial intelligence paradigms (MLP-ANN) and thermodynamic correlations

The solubility of four established anticancer drugs, including busulfan, decitabine, temozolomide, and tamoxifen, has already been documented in supercritical CO2 (scCO2) to develop advanced formulations. In the current research, their supercritical solubility was theoretically evaluated by some empirical models, such as PR and SRK (cubic equations of state), UNIQUAC and modified Wilson’s models (expanded liquid models), and an artificial intelligent paradigm. The critical properties of these drugs were calculated using approved group contribution methods. Among the empirical models, the Belghait model provided the best fit to the experimental data, yielding an AARD% value within the range of 5.45–9.37 and an approximate Radj value of 0.990. Comparing SRK and PR models, the SRK model exhibited higher accuracy in correlating solubility data of all the mentioned drugs, achieving mean AARD% and Radj values in the range of 9.32–15.37 and 0.98–0.996, respectively. The UNIQUAC model outperformed the modified Wilson model in accurately correlating the solubility data for all the specified drugs, achieving mean AARD% values of 5.04–6.48 and Radj values in the range of 0.981 to 0.991. Moreover, the artificial neural network model offered the highest accuracy in predicting the solubility of the desired drugs. Notably, approximately 99.7% of its predicted data can fit with the experimental data, indicating its extremely good performance.

Eye-Gaze-Guided Vision Transformer for Rectifying Shortcut Learning.

Learning harmful shortcuts such as spurious correlations and biases prevents deep neural networks from learning meaningful and useful representations, thus jeopardizing the generalizability and interpretability of the learned representation. The situation becomes even more serious in medical image analysis, where the clinical data are limited and scarce while the reliability, generalizability and transparency of the learned model are highly required. To rectify the harmful shortcuts in medical imaging applications, in this paper, we propose a novel eye-gaze-guided vision transformer (EG-ViT) model which infuses the visual attention from radiologists to proactively guide the vision transformer (ViT) model to focus on regions with potential pathology rather than spurious correlations. To do so, the EG-ViT model takes the masked image patches that are within the radiologists' interest as input while has an additional residual connection to the last encoder layer to maintain the interactions of all patches. The experiments on two medical imaging datasets demonstrate that the proposed EG-ViT model can effectively rectify the harmful shortcut learning and improve the interpretability of the model. Meanwhile, infusing the experts' domain knowledge can also improve the large-scale ViT model's performance over all compared baseline methods with limited samples available. In general, EG-ViT takes the advantages of powerful deep neural networks while rectifies the harmful shortcut learning with human expert's prior knowledge. This work also opens new avenues for advancing current artificial intelligence paradigms by infusing human intelligence.

Intelligent Decision Support for Energy Management: A Methodology for Tailored Explainability of Artificial Intelligence Analytics

This paper presents a novel development methodology for artificial intelligence (AI) analytics in energy management that focuses on tailored explainability to overcome the “black box” issue associated with AI analytics. Our approach addresses the fact that any given analytic service is to be used by different stakeholders, with different backgrounds, preferences, abilities, skills, and goals. Our methodology is aligned with the explainable artificial intelligence (XAI) paradigm and aims to enhance the interpretability of AI-empowered decision support systems (DSSs). Specifically, a clustering-based approach is adopted to customize the depth of explainability based on the specific needs of different user groups. This approach improves the accuracy and effectiveness of energy management analytics while promoting transparency and trust in the decision-making process. The methodology is structured around an iterative development lifecycle for an intelligent decision support system and includes several steps, such as stakeholder identification, an empirical study on usability and explainability, user clustering analysis, and the implementation of an XAI framework. The XAI framework comprises XAI clusters and local and global XAI, which facilitate higher adoption rates of the AI system and ensure responsible and safe deployment. The methodology is tested on a stacked neural network for an analytics service, which estimates energy savings from renovations, and aims to increase adoption rates and benefit the circular economy.

Focused Reservoir Fluid Sampling Uses Artificial Intelligence Technology

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 210091, “Optimizing Focused Reservoir Fluid Sampling Using a Deterministic Causation Artificial Intelligence Intuition Technology,” by Rabindra Chakraborty, Senslytics; Chengli Dong, Shell; and Hani Elshahawi, SPE, NoviDigiTech, et al. The paper has not been peer reviewed. _ Samples collected using wireline formation testing (WFT) provide vital information throughout the lifetime of a reservoir. Contaminated samples can lead to erroneous fluid analysis results with potentially huge economic consequences. A need exists for an application that can assist engineers in accurately inferring the state of fluid contamination. The complete paper describes the development of a WFT contamination-forewarning application based on a framework that advises real-time decisions regarding the state of fluid contamination and recommending changes that will help optimize the WFT operation. Focused Fluid Sampling During the past decade, focused fluid sampling has emerged as a viable alternative to conventional formation-fluid sampling. The method uses a special probing tool containing two distinct flow areas. The central (sample) inlet is connected to a sample line, often with an associated pump, while the peripheral (guard) ring is also associated with a dedicated pump. Focused sampling usually results in faster cleanup but comes with increased operational complexity. The flow rate of the sample and guard lines must be properly adjusted and synchronized for each case at hand, and the interpretation of downhole measurements on both sides is significantly more complex than for conventional sampling operations. The success of fluid-sampling operations is highly dependent on human interpretations of the real-time downhole sensor data. Focused sampling is carried out by manipulating the differential rates of sample and guard pumps to stimulate and accelerate sample cleanup. Various types of flow regimes observed during focused sampling operations may be classified as follows: - Idle: During such regimes, both sample and guard pumps are turned off. Thus, no cleanup takes place, and fluids in the tool’s sample lines that are detected by the downhole analyzers remain stagnant. - Commingled: During commingling, either the sample pump or the guard pump is running, but flow occurs through both the sample and guard areas combined. - Split flow: In such periods, both sample and guard pumps run simultaneously but at different flow rates, with fluids moving at different speeds through the sample (central) and guard (outer) areas. The physics of fluid flow in near-wellbore porous media are highly complex and challenging to compute analytically and difficult to numerically simulate at a relevant scale, especially in the context of real-time operations. The authors have developed what they term “intuition technology,” an artificial-intelligence paradigm that uses scientific learning as its fabric and mimics the human decision-making process by iterating hypotheses and weighing situational changes surrounding dynamics simultaneously while keeping an eye on possible asymptomatic behavior. The process involves harnessing experts’ subtle knowledge of fluid properties and change behavior as hypotheses and iterating them with the situational and time-series data to generate a scientific fabric that governs the cleanup. A contamination-forewarning application built on this intuition framework was able to interpret the fluid-cleanup state successfully in all examples examined as part of this pilot.

A Graph Neural Network Learning Approach to Optimize RIS-Assisted Federated Learning

Over-the-air federated learning (FL) is a promising privacy-preserving edge artificial intelligence paradigm, where over-the-air computation enables spectral-efficient model aggregation by achieving simultaneous communication and aggregation. However, due to limited transmit power, the performance of over-the-air FL is limited by the device with the worst channel condition toward the edge server. In this paper, we leverage reconfigurable intelligent surface (RIS) to mitigate the communication bottleneck of over-the-air FL and explicitly characterize the corresponding convergence upper bound. The convergence analysis illustrates the detrimental impact of the accumulated aggregation error over all rounds and inspires us to formulate a time-average transmission distortion minimization problem by jointly optimizing the transceiver and RIS phase-shifts. To reduce the computation complexity and enhance the model aggregation accuracy, we develop a graph neural network (GNN) based learning algorithm to directly map channel coefficients to the optimized network parameters. By exploiting permutation equivalence and invariance properties of graphs, the parameter dimension of the proposed algorithm is independent of the number of edge devices, which reduces the computational complexity and improves the algorithmic scalability. Simulations show that the proposed algorithm speeds up the computation by three orders of magnitude compared to the baselines, while achieving performance superiority and algorithmic robustness.