Fair Algorithm Research Articles

Analysis of the repercussions of Artificial Intelligence in the Personalization of the Virtual Educational Process in Higher Education Programs

This study examined how artificial intelligence (AI) has transformed the personalization of the virtual educational process in higher education programs. A systematic review of literature published between 2012 and 2023 was carried out, evaluating empirical studies, reports and review articles available in academic databases such as IEEE Xplore, SpringerLink and Google Scholar. Methods discussed include intelligent tutoring systems, learning analytics, and recommendation systems. The results showed that AI significantly improved the personalization of learning. Intelligent tutoring systems provide real-time adaptive feedback, adjusting content and pacing based on students' individual needs, improving their understanding and retention. Learning analytics helps identify student behavior patterns and predict academic issues, thereby facilitating timely interventions that help improve performance. Additionally, recommender systems personalize study materials based on student preferences and progress, thereby optimizing the educational experience. However, significant challenges have been identified, such as the need to protect data privacy and mitigate algorithmic biases that can affect the fairness and efficiency of these systems. In conclusion, the integration of AI into virtual higher education has enhanced the personalization of learning, improving both student satisfaction and academic performance. However, there is a need to continue to focus on developing ethical and equitable AI systems to address identified issues and maximize educational benefits

Addressing Algorithm Materiality and Bias Through Art: Exploring Social Media–Augmented Reality Features

Addressing Algorithm Materiality and Bias Through Art: Exploring Social Media–Augmented Reality Features

Analyzing the European institutional response to ethical and regulatory challenges of artificial intelligence in addressing discriminatory bias.

The European Union and some of its institutions have taken significant steps to address the challenges posed by the development and use of Artificial Intelligence (AI) in various contexts. The ubiquity of AI applications in everyday life, affecting both citizens and professionals, has made AI a common topic of discussion. However, as is evident from the documents analyzed here, concerns have been raised about the possible negative social consequences of AI, in particular discriminatory bias, making it a particularly relevant issue if people-centred, rights-based AI is to be implemented. This article aims to examine the challenges of defining, identifying and mitigating discriminatory bias in AI systems from two perspectives: (1) to conduct an ethical and normative review of European Commission documents from the last 8 years (from GDPR to AI Act regulation); and (2) to expose recommendations for key stakeholders, including designers, end-users and public authorities, to minimize/mitigate this risk. The document review was carried out on 21 EU regulatory and ethical guidelines in the field of AI, from which 152 measures were extracted, differentiated between design, governance and organizational measures. It has also been observed that there is no clear conceptual framework on the issue at the European level, showing a clear problem in providing definitions of algorithmic bias and discrimination, but not in assessing their potential negative impact on individuals. Secondly, these gaps may affect the concreteness and detail of the possible mitigation/minimization measures proposed and, subsequently, their application in different contexts. Finally, the last section of this paper presents a brief discussion and conclusions on possible issues related to the implementation of the measures extracted and certain limitations of the study.

Assessing racial bias in healthcare predictive models: Practical lessons from an empirical evaluation of 30-day hospital readmission models

ObjectiveDespite increased availability of methodologies to identify algorithmic bias, the operationalization of bias evaluation for healthcare predictive models is still limited. Therefore, this study proposes a process for bias evaluation through an empirical assessment of common hospital readmission models. The process includes selecting bias measures, interpretation, determining disparity impact and potential mitigations. MethodsThis retrospective analysis evaluated racial bias of four common models predicting 30-day unplanned readmission (i.e., LACE Index, HOSPITAL Score, and the CMS readmission measure applied as is and retrained). The models were assessed using 2.4 million adult inpatient discharges in Maryland from 2016 to 2019. Fairness metrics that are model-agnostic, easy to compute, and interpretable were implemented and apprised to select the most appropriate bias measures. The impact of changing model’s risk thresholds on these measures was further assessed to guide the selection of optimal thresholds to control and mitigate bias. ResultsFour bias measures were selected for the predictive task: zero-one-loss difference, false negative rate (FNR) parity, false positive rate (FPR) parity, and generalized entropy index. Based on these measures, the HOSPITAL score and the retrained CMS measure demonstrated the lowest racial bias. White patients showed a higher FNR while Black patients resulted in a higher FPR and zero-one-loss. As the models’ risk threshold changed, trade-offs between models’ fairness and overall performance were observed, and the assessment showed all models’ default thresholds were reasonable for balancing accuracy and bias. ConclusionsThis study proposes an Applied Framework to Assess Fairness of Predictive Models (AFAFPM) and demonstrates the process using 30-day hospital readmission model as the example. It suggests the feasibility of applying algorithmic bias assessment to determine optimized risk thresholds so that predictive models can be used more equitably and accurately. It is evident that a combination of qualitative and quantitative methods and a multidisciplinary team are necessary to identify, understand and respond to algorithm bias in real-world healthcare settings. Users should also apply multiple bias measures to ensure a more comprehensive, tailored, and balanced view. The results of bias measures, however, must be interpreted with caution and consider the larger operational, clinical, and policy context.

Potential application of artificial intelligence in cancer therapy.

This review underscores the critical role and challenges associated with the widespread adoption of artificial intelligence in cancer care to enhance disease management, streamline clinical processes, optimize data retrieval of health information, and generate and synthesize evidence. Advancements in artificial intelligence models and the development of digital biomarkers and diagnostics are applicable across the cancer continuum from early detection to survivorship care. Additionally, generative artificial intelligence has promised to streamline clinical documentation and patient communications, generate structured data for clinical trial matching, automate cancer registries, and facilitate advanced clinical decision support. Widespread adoption of artificial intelligence has been slow because of concerns about data diversity and data shift, model reliability and algorithm bias, legal oversight, and high information technology and infrastructure costs. Artificial intelligence models have significant potential to transform cancer care. Efforts are underway to deploy artificial intelligence models in the cancer practice, evaluate their clinical impact, and enhance their fairness and explainability. Standardized guidelines for the ethical integration of artificial intelligence models in cancer care pathways and clinical operations are needed. Clear governance and oversight will be necessary to gain trust in artificial intelligence-assisted cancer care by clinicians, scientists, and patients.

Deep learning-based prediction of one-year mortality in Finland is an accurate but unfair aging marker

Short-term mortality risk, which is indicative of individual frailty, serves as a marker for aging. Previous age clocks focused on predicting either chronological age or longer-term mortality. Aging clocks predicting short-term mortality are lacking and their algorithmic fairness remains unexamined. We developed a deep learning model to predict 1-year mortality using nationwide longitudinal data from the Finnish population (FinRegistry; n = 5.4 million), incorporating more than 8,000 features spanning up to 50 years. We achieved an area under the curve (AUC) of 0.944, outperforming a baseline model that included only age and sex (AUC = 0.897). The model generalized well to different causes of death (AUC > 0.800 for 45 of 50 causes), including coronavirus disease 2019, which was absent in the training data. Performance varied among demographics, with young females exhibiting the best and older males the worst results. Extensive prediction fairness analyses highlighted disparities among disadvantaged groups, posing challenges to equitable integration into public health interventions. Our model accurately identified short-term mortality risk, potentially serving as a population-wide aging marker.

ПРАВОВАЯ АРХИТЕКТУРА ВЗАИМОСВЯЗИ ИСКУССТВЕННОГО ИНТЕЛЛЕКТА И ЗАЩИТЫ ПРАВ ПАЦИЕНТОВ ПРИ МЕДИЦИНСКОЙ ДИАГНОСТИКЕ

The study of the legal architecture of the relationship between artificial intelligence and patient protection in medical diagnostics includes a brief overview of the current state of application of artificial intelligence in diagnostic medicine and an analysis of its potential benefits and risks. This article reviews the current regulations governing the use of artificial intelligence in medical diagnostics, including laws, guidelines and best practices, as well as the role of health care regulators and personal data protection. Special attention is paid to the ethical aspects of using artificial intelligence in diagnostics, such as: ensuring patient privacy, protecting medical data, preventing algorithmic bias, and ensuring transparency in the diagnostic decision-making process. Specific examples of implementation of artificial intelligence in diagnostic practice are presented, illustrating both opportunities and challenges associated with this technology, including analysis of successful and unsuccessful cases of implementation. The study describes promising developments in the field of AI diagnostics, including new technologies and trends, and forecasts the evolution of the legal framework in response to these innovations. Finally, it summarizes the results of the study and presents recommendations for improving the legal regulation of artificial intelligence in medical diagnostics, taking into account the need to protect the rights and interests of patients.

Navigating the ethical terrain of AI in higher education: Strategies for mitigating bias and promoting fairness

Artificial intelligence (AI) and machine learning (ML) are transforming higher education by enhancing personalized learning and academic support, yet they pose significant ethical challenges, particularly in terms of inherent biases. This review critically examines the integration of AI in higher education, underscoring the dual aspects of its potential to innovate educational paradigms and the essential need to address ethical implications to avoid perpetuating existing inequalities. The researchers employed a methodological approach that analyzed case studies and literature as primary data collection methods, focusing on strategies to mitigate biases through technical solutions, diverse datasets, and strict adherence to ethical guidelines. Their findings indicate that establishing an ethical AI environment in higher education is imperative and involves comprehensive efforts across policy regulation, governance, and education. The study emphasizes the significance of interdisciplinary collaboration in addressing the complexities of AI bias, highlighting how policy, regulation, governance, and education play pivotal roles in creating an ethical AI framework. Ultimately, the paper advocates for continuous vigilance and proactive strategies to ensure that AI contributes positively to educational settings, stressing the need for robust frameworks that integrate ethical considerations throughout the lifecycle of AI systems to ensure their responsible and equitable use.

Artificial intelligence and machine learning's role in sepsis-associated acute kidney injury.

Sepsis-associated acute kidney injury (SA-AKI) is a serious complication in critically ill patients, resulting in higher mortality, morbidity, and cost. The intricate pathophysiology of SA-AKI requires vigilant clinical monitoring and appropriate, prompt intervention. While traditional statistical analyses have identified severe risk factors for SA-AKI, the results have been inconsistent across studies. This has led to growing interest in leveraging artificial intelligence (AI) and machine learning (ML) to predict SA-AKI better. ML can uncover complex patterns beyond human discernment by analyzing vast datasets. Supervised learning models like XGBoost and RNN-LSTM have proven remarkably accurate at predicting SA-AKI onset and subsequent mortality, often surpassing traditional risk scores. Meanwhile, unsupervised learning reveals clinically relevant sub-phenotypes among diverse SA-AKI patients, enabling more tailored care. In addition, it potentially optimizes sepsis treatment to prevent SA-AKI through continual refinement based on patient outcomes. However, utilizing AI/ML presents ethical and practical challenges regarding data privacy, algorithmic biases, and regulatory compliance. AI/ML allows early risk detection, personalized management, optimal treatment strategies, and collaborative learning for SA-AKI management. Future directions include real-time patient monitoring, simulated data generation, and predictive algorithms for timely interventions. However, a smooth transition to clinical practice demands continuous model enhancements and rigorous regulatory oversight. In this article, we outlined the conventional methods used to address SA-AKI and explore how AI and ML can be applied to diagnose and manage SA-AKI, highlighting their potential to revolutionize SA-AKI care.

FairGAT: Fairness-Aware Graph Attention Networks

Graphs can facilitate modeling various complex systems such as gene networks and power grids as well as analyzing the underlying relations within them. Learning over graphs has recently attracted increasing attention, particularly graph neural network (GNN)–based solutions, among which graph attention networks (GATs) have become one of the most widely utilized neural network structures for graph-based tasks. Although it is shown that the use of graph structures in learning results in the amplification of algorithmic bias, the influence of the attention design in GATs on algorithmic bias has not been investigated. Motivated by this, the present study first carries out a theoretical analysis in order to demonstrate the sources of algorithmic bias in GAT-based learning for node classification. Then, a novel algorithm, FairGAT, which leverages a fairness-aware attention design, is developed based on the theoretical findings. Experimental results on real-world networks demonstrate that FairGAT improves group fairness measures while also providing comparable utility to the fairness-aware baselines for node classification and link prediction.

Embed systemic equity throughout industrial ecology applications: How to address machine learning unfairness and bias

AbstractRecent calls have been made for equity tools and frameworks to be integrated throughout the research and design life cycle —from conception to implementation—with an emphasis on reducing inequity in artificial intelligence (AI) and machine learning (ML) applications. Simply stating that equity should be integrated throughout, however, leaves much to be desired as industrial ecology (IE) researchers, practitioners, and decision‐makers attempt to employ equitable practices. In this forum piece, we use a critical review approach to explain how socioecological inequities emerge in ML applications across their life cycle stages by leveraging the food system. We exemplify the use of a comprehensive questionnaire to delineate unfair ML bias across data bias, algorithmic bias, and selection and deployment bias categories. Finally, we provide consolidated guidance and tailored strategies to help address AI/ML unfair bias and inequity in IE applications. Specifically, the guidance and tools help to address sensitivity, reliability, and uncertainty challenges. There is also discussion on how bias and inequity in AI/ML affect other IE research and design domains, besides the food system—such as living labs and circularity. We conclude with an explanation of the future directions IE should take to address unfair bias and inequity in AI/ML. Last, we call for systemic equity to be embedded throughout IE applications to fundamentally understand domain‐specific socioecological inequities, identify potential unfairness in ML, and select mitigation strategies in a manner that translates across different research domains.

TOWARDS ETHICAL AND SUSTAINABLE DEVELOPMENT IN BUSINESS MANAGEMENT WITH AI

This article considers into the ethical considerations surrounding the integration of artificial intelligence (AI) in business management. As AI becomes more prevalent in decision-making and operations, businesses encounter numerous ethical challenges. These include concerns about algorithmic bias, transparency, workforce implications, and data privacy. The ethical implications of AI in business are complex and far-reaching. This thorough analysis examines real-world examples, ethical frameworks, and best practices to help businesses responsibly navigate these challenges. By proactively addressing these ethical dilemmas, businesses can harness the advantages of AI while maintaining ethical standards and societal values.

Automated construction of effective potential via algorithmic implicit bias

We introduce a novel approach for decomposing and learning every scale of a given multiscale objective function in Rd, where d⩾1. This approach leverages a recently demonstrated implicit bias of the optimization method of gradient descent [44], which enables the automatic generation of data that nearly follow Gibbs distribution with an effective potential at any desired scale. One application of this automated effective potential modeling is to construct reduced-order models. For instance, a deterministic surrogate Hamiltonian model can be developed to substantially soften the stiffness that bottlenecks the simulation, while maintaining the accuracy of phase portraits at the scale of interest. Similarly, a stochastic surrogate model can be constructed at a desired scale, such that both its equilibrium and out-of-equilibrium behaviors (characterized by auto-correlation function and mean path) align with those of a damped mechanical system with the original multiscale function being its potential. The robustness and efficiency of our proposed approach in multi-dimensional scenarios have been demonstrated through a series of numerical experiments. A by-product of our development is a method for anisotropic noise estimation and calibration. More precisely, Langevin model of stochastic mechanical systems may not have isotropic noise in practice, and we provide a systematic algorithm to quantify its covariance matrix without directly measuring the noise. In this case, the system may not admit closed form expression of its invariant distribution either, but with this tool, we can design friction matrix appropriately to calibrate the system so that its invariant distribution has a closed form expression of Gibbs.

Pushing Boundaries: AI and Computer Science in the Era of Technological Revolution

The advent of Artificial Intelligence (AI) has heralded a transformative era in Computer Science, revolutionizing various facets of technology. This paper explores the profound impact of AI on the field of Computer Science, delving into its advancements, innovations, and the potential future trajectory of technology. The primary objective of this paper is to elucidate the significant role of AI in shaping the landscape of Computer Science. Through comprehensive research and analysis, it aims to provide insights into the evolution of AI, its applications, and the implications for future technological developments. This study employs a library research approach, gathering information from academic journals, conference proceedings, books, and reputable online sources. By synthesizing existing literature and scholarly works, it seeks to construct a comprehensive overview of the advancements and innovations in AI within the realm of Computer Science. The findings reveal a dynamic and rapidly evolving field driven by AI technologies. From machine learning algorithms to neural networks and deep learning models, AI has revolutionized data analysis, pattern recognition, and decision-making processes. Moreover, AI-driven applications such as natural language processing, computer vision, and robotics have reshaped various industries, including healthcare, finance, transportation, and manufacturing. However, alongside these advancements come ethical considerations, privacy concerns, and challenges related to algorithmic biases and societal implications. Looking ahead, the future of technology promises further integration of AI into various domains, leading to unprecedented opportunities and challenges in the realm of Computer Science

Gender Bias in AI

This work analyzes the gendered bias inherent in burgeoning Artificial Intelligence (AI) systems. Drawing from feminist scholarship on algorithmic bias, it argues that AI development reflects and amplifies existing patriarchal structures. The analysis highlights how AI applications, from chatbots reinforcing gendered stereotypes to art generators perpetuating a white male gaze, threaten to exacerbate social inequities

The AI Jungle Book!

The integration of Artificial Intelligence (AI) into various sectors is both revolutionary and fraught with ethical, social, and regulatory challenges. This paper explores the complex landscape of AI governance, emphasizing the critical balance between innovation and responsibility. By examining the multifaceted impacts of AI, including algorithmic bias, privacy concerns, safety hazards, and ethical considerations, this review highlights the urgent need for a comprehensive framework of harmony. Through a synthesis of key literature and thought leadership from notable scholars, this study underscores the importance of ethical stewardship, collaborative policymaking, and international cooperation in navigating the AI jungle.

Ethical Considerations in the Use of Artificial Intelligence and Machine Learning in Health Care: A Comprehensive Review.

Artificial intelligence (AI) and machine learning (ML) technologies are revolutionizing health care by offering unprecedented opportunities to enhance patient care, optimize clinical workflows, and advance medical research. However, the integration of AI and ML into healthcare systems raises significant ethical considerations that must be carefully addressed to ensure responsible and equitable deployment. This comprehensive review explored the multifaceted ethical considerations surrounding the use of AI and ML in health care, including privacy and data security, algorithmic bias, transparency, clinical validation, and professional responsibility. By critically examining these ethical dimensions, stakeholders can navigate the ethical complexities of AI and ML integration in health care, while safeguarding patient welfare and upholding ethical principles. By embracing ethical best practices and fostering collaboration across interdisciplinary teams, the healthcare community can harness the full potential of AI and ML technologies to usher in a new era of personalized data-driven health care that prioritizes patient well-being and equity.

Enhancing retinal disease diagnosis through AI: Evaluating performance, ethical considerations, and clinical implementation

BackgroundThe research problem addresses the need for accurate and efficient detection of retina diseases using artificial intelligence (AI) technologies. The specific aim is to evaluate the performance, ethical considerations, and clinical implementation of AI-driven retina disease detection systems. MethodsThis study is a systematic review. Data sources assessed included various electronic databases searched up to July 31, 2023. The prespecified criteria for study inclusion were studies involving AI algorithms for retina disease detection, including those focused on diabetic retinopathy, age-related macular degeneration, and glaucoma. Participant eligibility criteria encompassed human subjects of all ages, and the interventions assessed were AI-based diagnostic tools compared to traditional diagnostic methods. Only randomized controlled trials and observational studies set in clinical environments were included, covering a time span from the inception of AI technology. FindingsThe search identified 145 studies, of which 61 met the inclusion criteria and were eligible for analysis. The narrative summary of findings indicated that AI algorithms generally demonstrated high accuracy, sensitivity, and specificity in detecting retinal diseases. Deep learning algorithms showed a sensitivity of 90 % and specificity of 98 % for diabetic retinopathy detection. However, several studies highlighted concerns about algorithmic bias, data privacy, and the need for diverse and representative datasets to ensure generalizability across different populations. InterpretationThe AI-driven retina disease detection systems have significant potential to improve diagnostic accuracy and efficiency in clinical practice. Ethical considerations regarding patient privacy, the risk of algorithmic bias, and the challenges of integrating AI into existing healthcare workflows must be addressed. The study underscores the importance of ongoing validation, ethical scrutiny, and interdisciplinary collaboration to harness the benefits of AI while mitigating its risks, ensuring responsible and equitable implementation in clinical settings.

Surveying Racial Bias in Facial Recognition: Balancing Datasets and Algorithmic Enhancements

Surveying Racial Bias in Facial Recognition: Balancing Datasets and Algorithmic Enhancements

Artificial Intelligence’s Opportunities and Challenges in Engineering Curricular Design: A Combined Review and Focus Group Study

This study explores the opportunities and challenges of integrating artificial intelligence (AI) into engineering education. Through a review of the literature and a qualitative focus group study, an assessment was made for the role of AI in personalizing learning, enhancing simulation engagement, providing real-time feedback, and preparing students for AI-integrated workplaces. The study emphasizes how AI may significantly improve educational experiences by making them more dynamic, interactive, and successful. It also draws attention to important issues, such as moral questions, algorithmic biases in AI, infrastructure constraints, the need for AI literacy training for educators, and a range of student perspectives on AI engineering education. The results support a systematic approach to AI integration, highlighting the necessity of cooperative efforts by educators, legislators, curriculum designers, and technologists in order to overcome these obstacles. The study makes the case that AI can transform engineering education by negotiating these challenges and providing students with the information and skills needed for the digital future, all the while assuring fair and moral access to technology-enhanced learning.