Computer Science Research Articles

Transfer learning for accurate description of atomic transport in Al-Cu melts.

Machine learning interatomic potentials (MLIPs) provide an optimal balance between accuracy and computational efficiency and allow studying problems that are hardly solvable by traditional methods. For metallic alloys, MLIPs are typically developed based on density functional theory with generalized gradient approximation (GGA) for the exchange-correlation functional. However, recent studies have shown that this standard protocol can be inaccurate for calculating the transport properties or phase diagrams of some metallic alloys. Thus, optimization of the choice of exchange-correlation functional and specific calculation parameters is needed. In this study, we address this issue for Al-Cu alloys, in which standard Perdew-Burke-Ernzerhof (PBE)-based MLIPs cannot accurately calculate the viscosity and melting temperatures at Cu-rich compositions. We have built MLIPs based on different exchange-correlation functionals, including meta-GGA, using a transfer learning strategy, which allows us to reduce the amount of training data by an order of magnitude compared to a standard approach. We show that r2SCAN- and PBEsol-based MLIPs provide much better accuracy in describing thermodynamic and transport properties of Al-Cu alloys. In particular, r2SCAN-based deep machine learning potential allows us to quantitatively reproduce the concentration dependence of dynamic viscosity. Our findings contribute to the development of MLIPs that provide quantum chemical accuracy, which is one of the most challenging problems in modern computational materials science.

Extending GPU-accelerated Gaussian integrals in the TeraChem software package to f type orbitals: Implementation and applications.

The increasing availability of graphics processing units (GPUs) for scientific computing has prompted interest in accelerating quantum chemical calculations through their use. However, the complexity of integral kernels for high angular momentum basis functions often limits the utility of GPU implementations with large basis sets or for metal containing systems. In this work, we report the implementation of f function support in the GPU-accelerated TeraChem software package through the development of efficient kernels for the evaluation of Hamiltonian integrals. The high efficiency of the resulting code is demonstrated through density functional theory (DFT) calculations on increasingly large organic molecules and transition metal complexes, as well as coupled cluster singles and doubles calculations on water clusters. Preliminary investigations into Ni(I) catalysis with DFT and the photochemistry of MnH(CH3) with complete active space self-consistent field are also carried out. Overall, our GPU-accelerated software appears to be well-suited for fast simulation of large transition metal containing systems, as well as organic molecules.

Necessary reforms in the Greek academic system

Motivated by our EU funded ERA chair projects, that require proposing reforms to yield the Greek academic system more competitive with a particular emphasis on fostering brain gain and preventing brain drain, we propose policy changes on a wide range of topics, based on our joint academic management experience inside and outside of Greece in conjunction with our diverse backgrounds in the humanities, computer science, and the life sciences. The proposed reforms can contribute to increase the competitiveness of Greece in research and development (R & D). We classify proposals by their associated costs and include a plethora of zero cost proposals that do have the potential to increase efficiency and net research time as well as to improve current working conditions in Greek academia. However, to yield Greece more competitive at the international level, reverse brain drain, and foster brain gain, substantial investments and increases of R&D expenditure are required which depend on political willingness and require a long term strategic development plan for Greece beyond being a tourist destination in the European periphery.

A scoping review of the literature on computational thinking in education in the Arab world

This study aims to identify the nature and the extent of recent research studies on computational thinking in the Arab educational systems. The analyzed studies were gathered from English and Arabic educational sources. The selected publications were peer-reviewed journal articles over the past ten years that dealt with computational thinking in education in the Arab world. The selected studies were analyzed based on language, publication date, educational setting, educational discipline, site, primary purpose, research method, sample, and adopted definition of computational thinking. Twenty-eight research articles were selected for the investigation. The analysis revealed that the majority of the examined research studies were published in the Arabic language after the year 2020 in K-12 educational settings. The majority of the examined studies were conducted in programming and computer science in a few Arab countries. In addition, the majority of the examined studies focused on how to develop students’ computational thinking levels using various interventions. The majority of the studies employed a limited number of participants and used different forms of experimental research designs. The studies adopted different forms of the definition of computational thinking. Based on the findings, recommendations for future studies were presented.

John Robots, Thurgood Martian, and the Syntax Monster: A New Argument Against AI Judges

Abstract This paper argues that an AI judge is conceptually undesirable and not just something that lies beyond the state of the art in computer science. In a nutshell, even if an AI system could accurately predict how a good human judge would decide a particular case, its prediction would be the product of correlations between such factors as patterns of syntax in bodies of legal texts. This approach of AI systems is insufficient for basing their output on the sort of rationales that are expected of valid judicial decisions in any desirable legal system. Thus, by their very nature, AI systems are incapable of providing valid legal decisions in any such system.

Object-oriented paradigm as a method for formalizing character properties in fictional universes (on the example of the character representation of Nekhlyudov in Leo Tolstoy’s novel Resurrection)

The article explores the possibility of applying the object-oriented paradigm (a term sourced from computer science) to formalize various implementations of characters in fictional universes. Examples are provided based on the implementations of the character of Nekhlyudov from Tolstoy’s novel Resurrection. Possible applications of the discussed approach are examined for analyzing the relationships between different implementations of the same character, as well as for generating new characters based on the individual preferences of the reader (or viewer, user).

Gerontologic Biostatistics and Data Science: Aging Research in the Era of Big Data.

Introduced in 2010, the sub-discipline of gerontologic biostatistics was conceptualized to address the specific challenges of analyzing data from clinical research studies involving older adults. Since then, the evolving technological landscape has led to a proliferation of advancements in biostatistics and other data sciences that have significantly influenced the practice of gerontologic research, including studies beyond the clinic. Data science is the field at the intersection of statistics and computer science, and although the term "data science" was not widely used in 2010, the field has quickly made palpable impacts on gerontologic research. In this Review in Depth, we describe multiple advancements of biostatistics and data science that have been particularly impactful. Moreover, we propose the sub-discipline of "gerontologic biostatistics and data science", or GBDS, which subsumes gerontologic biostatistics into a more encompassing practice. Prominent GBDS advancements that we discuss herein include cutting-edge methods in experimental design and causal inference, adaptations of machine learning, the rigorous quantification of deep phenotypic measurement, and analysis of high-dimensional -omics data. We additionally describe the need for integration of information from multiple studies and propose strategies to foster reproducibility, replicability, and open science. Lastly, we provide information on software resources for GBDS practitioners to apply these approaches to their own work and propose areas where further advancement is needed. The methodological topics reviewed here aim to enhance data-rich research on aging and foster the next generation of gerontologic researchers.

Fault-tolerant partition resolvability of cycle with chord.

In the realm of connected networks, distance-based parameters, particularly the partition dimension of graphs, have extensive applications across various fields, including chemistry and computer science. A notable variant of the partition dimension is the fault-tolerant resolving partition, which is critical in computer science for networking, optimization, and navigation tasks. In networking, fault-tolerant partitioning ensures robust communication pathways even in the event of network failures or disruptions. In optimization, it aids in developing efficient algorithms capable of withstanding errors or changes in input data. In navigation systems, fault-tolerant partitioning supports reliable route planning and navigation services under uncertain or dynamic conditions. This paper focuses on the fault-tolerant partition dimension within the specific context of the cycle with chord graphs, exploring its properties and implications for enhancing the robustness and reliability of networked systems.

Metaphorical meaning dynamics: Identifying patterns in the metaphorical evolution of English words using mathematical modeling techniques

Abstract Conceptual metaphor theory has been criticized due to its emphasis on concepts instead of words and its top-down direction of analysis. In response to these criticisms, this paper employs a new strategy, utilizing established mathematical modeling methods to allow a systematic, quantitative analysis of the entire dataset produced by the Mapping Metaphor project at the University of Glasgow. This dataset consists of 9609 words performing 18916 metaphorical mappings between 414 domains. The data is represented as a network consisting of 414 nodes, the domains, connected by shared words. Words are represented by groups of directed mappings between all domains in which they occur. This is made possible by the use of a directed hypergraph representation, a tool commonly used in discrete mathematics and various areas of computer science but not previously applied to the metaphorical meanings of words. Examining the dataset as a whole, rather than focusing on individual words or metaphors, allows global patterns of behavior to emerge from the data without pre-filtering or selection by the authors. Outcomes of the analysis relating to the distributions of source and target domains within the network, the growth mechanisms at work in the spread of metaphorical meanings and how these relate to existing concepts in CMT are discussed.

A Novel Noise-Aware Classical Optimizer for Variational Quantum Algorithms

A key component of variational quantum algorithms (VQAs) is the choice of classical optimizer employed to update the parameterization of an ansatz. It is well recognized that quantum algorithms will, for the foreseeable future, necessarily be run on noisy devices with limited fidelities. Thus, the evaluation of an objective function (e.g., the guiding function in the quantum approximate optimization algorithm (QAOA) or the expectation of the electronic Hamiltonian in variational quantum eigensolver (VQE)) required by a classical optimizer is subject not only to stochastic error from estimating an expected value but also to error resulting from intermittent hardware noise. Model-based derivative-free optimization methods have emerged as popular choices of a classical optimizer in the noisy VQA setting, based on empirical studies. However, these optimization methods were not explicitly designed with the consideration of noise. In this work we adapt recent developments from the “noise-aware numerical optimization” literature to these commonly used derivative-free model-based methods. We introduce the key defining characteristics of these novel noise-aware derivative-free model-based methods that separate them from standard model-based methods. We study an implementation of such noise-aware derivative-free model-based methods and compare its performance on demonstrative VQA simulations to classical solvers packaged in scikit-quant. History: Accepted by Giacomo Nannicini, Area Editor for Quantum Computing and Operations Research. Accepted for Special Issue. Funding: This material is based upon work supported by the U.S. Department of Energy, Office of Science, National Quantum Information Science Research Centers and the Office of Advanced Scientific Computing Research, Accelerated Research for Quantum Computing program under contract number DE-AC02-06CH11357. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2023.0177 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2023.0177 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .

A practitioner-centered policy roadmap for ethical computational social science in Germany, Austria, and Switzerland.

Background: Computational Social Science (CSS) utilizes large digital datasets and computational methods to study human behavior, raising ethical concerns about data privacy, informed consent, and potential misuse.Methods: This study employs a constructivist grounded theory approach, analyzing 15 in-depth interviews with CSS practitioners in Germany, Austria, and Switzerland. These countries share a European legal context regarding data privacy and hereby provide a comparable regulatory environment for examining ethical considerations.Results: Findings highlight key challenges in CSS research, including power imbalances with data providers, uncertainties around surveillance and data privacy (especially with longitudinal data), and limitations of current ethics frameworks. Researchers face tensions between established ethical principles and practical realities, often feeling disempowered and lacking support from ethics boards due to their limited CSS expertise. Regulatory ambiguity further discourages research due to fear of sanctions.Conclusions: To foster responsible CSS practices, this paper recommends establishing specialized ethics boards with CSS expertise. It also advocates for acknowledging CSS's unique nature in research policy by developing tailored data guidelines and providing legal certainty through clear guidelines. Grounding recommendations in practitioners' experiences, this study offers actionable steps to help enable ethical CSS research.

Analysing the psychological capital influence on academic performance in an e-learning environment

PurposeThe purpose of this study is to examine the evolving landscape of e-learning, which has become a transformative educational approach accelerated by technological advancements and the global impact of the COVID-19 pandemic. It aims to explore the adoption and impact of e-learning across diverse educational settings, focusing on its effectiveness, quality and potential challenges. Furthermore, this research delves into the often-overlooked role of psychological resources and capabilities, particularly Psychological Capital (PsyCap), in enhancing academic performance among university students engaged in e-learning. By investigating the influence of psychological resources and their intersection with e-learning, this study seeks to contribute to a deeper understanding of the factors that shape successful e-learning experiences and academic achievement.Design/methodology/approachIn this research methodology, data was collected from an e-learning institution, with a focus on students in various computer science semesters to ensure comprehensive representation. To minimize standard method bias, a random sampling technique was employed, and data collection was conducted with the support of locally hired research associates. Participant confidentiality and anonymity were carefully preserved, and ethical approval was obtained. The study began with demographic data collection and an assessment of PsyCap dimensions. Most measurements were self-reported, except for GPA, retrieved from the institute directly. The study initially involved 468 students, but only 213 provided complete responses, resulting in a 46% response rate. Demographic data included age, marital status and gender. The sample featured diverse generational representation, with 58% from Generation Z, 13% Millennials, 22% Generation X and 6% Baby Boomers. This comprehensive data will help analyse generational influences on e-learning outcomes.FindingsThe study’s findings underscore the significance of self-efficacy, hope, resilience and optimism in e-learning success. Higher self-efficacy positively impacts student engagement, aligning with previous research. PsyCap, which includes these traits, proves relevant in educational settings. Resilience is notably beneficial, aiding students in overcoming challenges and bolstering their self-belief. Hope enhances problem-solving and adaptability, while optimism fosters a proactive attitude and perseverance, both vital for academic excellence. These insights have broad implications for e-learning practices, emphasizing the need to nurture psychological resources. Incorporating PsyCap-based interventions can enhance the educational experience, promoting student success.Research limitations/implicationsThis study has certain limitations that should be considered. It focused exclusively on computer science students, which may limit the generalizability of the findings. Future research should encompass a wider range of academic disciplines for broader applicability. The study was conducted within a specific cultural and regional context, emphasizing a collectivist culture in an Asian setting. To enhance the applicability of the results, it is crucial to explore different regions and cultural contexts. While the study controlled for generational cohort effects on academic performance, further investigation is warranted to understand how different generational cohorts perceive e-learning. Additionally, the study suggests examining how psychological resources influence students' perceptions of e-learning as a stressor or motivator. Furthermore, a comprehensive study comparing the impact of PsyCap on academic performance in both e-learning and traditional education, involving diverse samples and various cultural settings, is needed to provide a more comprehensive understanding.Originality/valueThis research contributes by thoroughly examining the impact of e-learning in diverse educational settings, focusing on its effectiveness, quality and potential challenges. A novel aspect is the exploration of the often-overlooked role of PsyCap in enhancing academic performance among university students engaged in e-learning. This sheds light on the intersection of psychological resources and e-learning. Additionally, the study’s rigorous research methodology underscores its commitment to ethical and responsible research conduct. The research also presents valuable demographic data on generational cohorts and gender, offering insights into how these factors influence e-learning outcomes. These original contributions collectively enhance our understanding of the multifaceted dynamics of e-learning and the pivotal role of psychological resources in academic success.

An approach for regime-switching stochastic control problems with memory and terminal conditions

In this research article, we focus on a stochastic optimal control problem with two types of terminal constraints. These specific conditions provide real-valued and stochastic Lagrange multipliers. Our model evolves according to a Markov regime-switching jump diffusion model with memory. In this context, the memory is represented by a Stochastic Differential Delay Equation. We present two theorems for each constraint within the general formulation of stochastic optimal control theory in a Lagrangian environment. We approach to this task from a theoretical perspective and provide mild technical assumptions, which make our theorems applicable for a broad class of stochastic control problems as well as for a wide range of disciplines such as engineering, biology, operations research, medicine, computer science and economics. In this work, we apply Stochastic Maximum Principle to demonstrate an optimal dividend policy corresponding to a time-delayed wealth process of a company. Moreover, we determine the real-valued Lagrange multiplier of this control problem explicitly.

"I'm in the Bluesky Tonight": Insights from a year worth of social data.

Pollution of online social spaces caused by rampaging d/misinformation is a growing societal concern. However, recent decisions to reduce access to social media APIs are causing a shortage of publicly available, recent, social media data, thus hindering the advancement of computational social science as a whole. We present a large, high-coverage dataset of social interactions and user-generated content from Bluesky Social to address this pressing issue. The dataset contains the complete post history of over 4M users (81% of all registered accounts), totalling 235M posts. We also make available social data covering follow, comment, repost, and quote interactions. Since Bluesky allows users to create and like feed generators (i.e., content recommendation algorithms), we also release the full output of several popular algorithms available on the platform, along with their timestamped "like" interactions. This dataset allows novel analysis of online behavior and human-machine engagement patterns. Notably, it provides ground-truth data for studying the effects of content exposure and self-selection and performing content virality and diffusion analysis.

Navigating the Privacy Paradox : Balancing AI Advancement and Data Protection in the Digital Age

The rapid advancement of artificial intelligence (AI) technologies has ushered in an era of unprecedented data utilization, raising critical concerns about individual privacy. This article examines the complex interplay between AI development and data protection, exploring the challenges of balancing the need for large-scale data processing with the imperative to safeguard personal information. We analyze current regulatory frameworks, including GDPR and CCPA, and their efficacy in addressing AI-specific privacy issues. The article also evaluates technical measures such as differential privacy, federated learning, privacy-preserving machine learning algorithms, and organizational best practices for responsible data management. By synthesizing computer science, law, and ethics perspectives, we propose a multidisciplinary approach to fostering AI innovation while robustly protecting individual privacy rights. Our findings underscore the need for adaptive regulations, enhanced technical safeguards, and increased collaboration between stakeholders to navigate the evolving landscape of AI and privacy in the digital age.

An interview with Professor Benjamin Becker: understanding our brain and mental disorders requires collaboration across all disciplines

Abstract From July 20th to 22nd, 2024, the ISMRM Endorsed Workshop on MR for Psychiatry was held in Chengdu City, China. This prestigious event attracted numerous academic elites worldwide, and Professor Benjamin Becker from the University of Hong Kong was invited. On the morning of July 20, during the “Advances in MR Technology” session, Professor Becker delivered an engaging lecture entitled “Novel approaches to precision MRI-imaging of human emotion”. His presentation was met with great enthusiasm and sparked lively discussions among the participants. Following the conference, the Psychoradiology journal interviewed Professor Benjamin Becker. In the interview, Benjamin emphasized the significant role of interdisciplinary collaboration, spanning various fields including psychology, neuroscience, clinical medicine, biomedical engineering, and computer science. Professor Becker firmly believed that such collaboration was crucial for a deeper understanding of the brain and psychiatric disorders. Additionally, he highly valued the importance of international cooperation, especially in addressing global mental health issues and challenges related to psychiatric disorders.

PPSZ for General k-SAT and CSP—Making Hertli’s Analysis Simpler and 3-SAT Faster

The currently fastest known algorithm for k-SAT is PPSZ, named afterits inventors (Paturi et al. in J ACM 52(3):337-364, 2005. http://dx.doi.org/10.1145/1066100.1066101). Analyzing its running timeis much easier for input formulas with a unique satisfying assignment.In this paper, we achieve three goals. First, we simplify the analysisof Hertli (in 2011 IEEE 52nd Annual Symposium on Foundations ofComputer Science-FOCS 2011, Los Alamitos, 2011) for input formulaswith multiple satisfying assignments. Second, we show a “liftingresult”: if you improve PPSZ for k-CNF formulas with a unique satisfyingassignment, you will immediately get a (weaker) improvement forgeneral k-CNF formulas. In combination this with results by Hansen etal. (in Charikar and Cohen (ed) Proceedings of the 51st Annual ACMSIGACT Symposium on Theory of Computing, 2019) and Scheder (in62nd IEEE Annual Symposium on Foundations of Computer Science,2021), who all prove improved time bounds for Unique-k-SAT, this givesimproved bounds for general k-SAT. We also generalize our results tothe domain of Constraint Satisfaction Problems, i.e., satisfiability withmore than two truth values.

Evolving Computer Science Education Research in Teaching and Learning: A Bibliometric Analysis

This study performed a bibliometric analysis of computer science education (CSE) publications in teaching and learning. The research trend in computer science education was analysed using publications from the Scopus-indexed database between 1970 and 2021. This study aims to summarize this information to assist three groups of stakeholders: educators, researchers and policymakers. Our review goal to document and analyse the trends in CSE research publications for teaching and learning. A total of 800 publications were derived for further assessment based on computer science education in the article title with multiple tools. We used, Microsoft Excel, VOSviewer and Harzing’s Publish or Perish were utilized for frequency analysis data visualization and citation metrics and assessment. The study outcomes were documented with standard bibliometric indicators: publication growth, authorship trends, collaboration, prolific authors, national contribution, most active establishments, top-cited articles, preferred journals, intellectual structure and topical trends. Resultantly, CSE research publications have perpetually grown for the past 51 years since 1970 with the United States highlighting the largest study contributor, followed by Germany and the United Kingdom. The ACM SIGCSE Bulletin published the highest number of CSE research publications. The analysis of the cluster on topical trends indicates that curriculum design and implementation (e.g., assessment, broadening participation), elements of teaching and learning (e.g., constructivism, e-learning, game-based learning) and other areas of interest comprise pedagogical approach (e.g., computational thinking, k-12 education, motivation). Based on the research results, we expand on the findings and propose how authors might enhance the CSE function in curriculum design, implementation and shifts could eventually substitute past study clusters.

Human-Machine Engagement (HME): Conceptualization, Typology of Forms, Antecedents, and Consequences

Artificial intelligence (AI) applications in customer-facing settings are growing rapidly. The general shift toward robot- and AI-powered services prompts a reshaping of customer engagement, bringing machines into engagement conceptualizations. In this paper, we build on service research around engagement and AI, incorporating computer science, and socio-technical systems perspective to conceptualize human-machine engagement (HME), offering a typology and nomological network of antecedents and consequences. Through three empirical studies, we develop a typology of four distinct forms of HME (informative, experimenting, praising, apprehensive), which differ in valence and intensity, underpinned by both emotional (excitement) and cognitive (concern, advocacy) drivers. We offer empirical evidence which reveals how these HME forms lead to different cognitive and personality-related outcomes for other users (perceived value of HME, perceived risk, affinity with HME) and service providers (willingness to implement in services, perceived value of HME). We also reveal how outcomes for service providers vary with the presence and absence of competitor pressure. Our findings broaden the scope of engagement research to include non-human actors and suggest both strategic and tactical guidance to service providers currently using and/or seeking to use generative AI (GenAI) in services alongside an agenda to direct future studies on HME.

Improving Retention and Success of African American Students in Computer Science: A Mixed-Method Case Study at an HBCU

This case study introduces the STARS (Supporting Talented African American Undergraduates for Retention and Success) project, designed to foster the retention and success of academically talented African American computer science students from low-income backgrounds at Historically Black Colleges and Universities (HBCUs) in the U.S. The STARS program employs a holistic approach, integrating four primary pillars of support: academic, social, career, and financial. Specific support provided includes near-peer mentoring, technical skill development seminars, undergraduate research, and high school outreach activities. To explore the program’s effectiveness and areas of improvement, a mixed-method evaluation study was conducted, collecting data through surveys, observations, individual interviews, and focus group interviews. The findings revealed that the STARS program contributed to high levels of retention among its scholars, and the mentoring program provided valuable networking opportunities. The study suggests that the program’s comprehensive approach, tailored to scholars’ needs, and combined with a culturally affirming learning environment, facilitates the retention and success of talented African American students in computer science.