Gurmit Singh, Simant Kamal Dutta, Rajesh Singh
et al.
Currently, industry 4.0 technologies have revolutionised every field for digital transformation, real-time monitoring, and innovative infrastructure. Real-time insights across war fighting and real-time decision support to individual and realistic training scenarios for soldiers are all positive outcomes of assimilating Industry 4.0 into Defence. In this study, we have presented a detailed discussion on the significance of Industry 4.0 technologies for Defence from the perspective of innovation. During the analysis, it was concluded that artificial intelligence (AI), robots, drones, augmented reality, simulation, additive manufacturing, big data analytics, and Internet of Things (IoT) are Industry 4.0, technologies that are used in Defence for trajectory analysis of enemy zones, smart bullets, anti-drone Defence systems, defining the target range of missiles, target classification, rapid prototyping of deployed military equipment, extracting injured personnel, and remote firing. This study proposes an industrial Defence sustainability and innovation framework to advance understanding of how industry 4.0 can enable innovation.
History of scholarship and learning. The humanities, Social sciences (General)
Decentralized learning provides a scalable alternative to parameter-server-based training, yet its performance is often hindered by limited peer-to-peer communication. In this paper, we study how communication should be scheduled over time, including determining when and how frequently devices synchronize. Counterintuitive empirical results show that concentrating communication budgets in the later stages of decentralized training remarkably improves global test performance. Surprisingly, we uncover that fully connected communication at the final step, implemented by a single global merging, can significantly improve the performance of decentralized learning under high data heterogeneity. Our theoretical contributions, which explain these phenomena, are the first to establish that the globally merged model of decentralized SGD can match the convergence rate of parallel SGD. Technically, we reinterpret part of the discrepancy among local models, which were previously considered as detrimental noise, as constructive components essential for matching this rate. This work provides evidence that decentralized learning is able to generalize under high data heterogeneity and limited communication, while offering broad new avenues for model merging research.
Christos Sgouropoulos, Christos Nikou, Stefanos Vlachos
et al.
Few-shot learning has emerged as a powerful paradigm for training models with limited labeled data, addressing challenges in scenarios where large-scale annotation is impractical. While extensive research has been conducted in the image domain, few-shot learning in audio classification remains relatively underexplored. In this work, we investigate the effect of integrating supervised contrastive loss into prototypical few shot training for audio classification. In detail, we demonstrate that angular loss further improves the performance compared to the standard contrastive loss. Our method leverages SpecAugment followed by a self-attention mechanism to encapsulate diverse information of augmented input versions into one unified embedding. We evaluate our approach on MetaAudio, a benchmark including five datasets with predefined splits, standardized preprocessing, and a comprehensive set of few-shot learning models for comparison. The proposed approach achieves state-of-the-art performance in a 5-way, 5-shot setting.
Pemahaman terhadap tipe kepribadian menjadi mutlak pada kondisi digitalisasi dan hybrid working. Tipe kepribadian yang umum dikenal saat ini adalah introver dan ekstrover. Organisasi yang tidak mampu memahami tipe kepribadian karyawan, akan berdampak pada penurunan motivasi dan kinerja karyawan. Salah satu cara mengklasifikasikan tipe kepribadian pegawai adalah dengan pendekatan machine learning. Evaluasi terhadap beberapa hasil pendekatan machine learning, akan memberikan model dengan kinerja terbaik yang mampu mengklasifikasikan tipe kepribadian. Model Naïve Bayes menjadi model terbaik pada klasfikasi tipe kepribadian ini dengan nilai accuracy sebesar 93,41%, lebih tinggi dibandingkan model lainnya. Penelitian ini diharapkan menambah wawasan ilmu pengetahuan pada human resources analitik dan memberikan informasi klasifikasi tipe kepribadian karyawan bagi organisasi.
Exploring the rise of open scholarship in the digital era and its transformational impact on how knowledge is created, shared, and accessed, this open access book offers new insights on the history, development, and future directions of openness in the humanities and identifies key drivers, opportunities, and challenges. The concept of open research is reconfiguring scholarly communication across all disciplines, changing how understandings are produced through more accessible, participatory, ethical, and transparent approaches, reaching and involving far broader and more diverse publics. Considering multiple stakeholder perspectives, Arthur and Hearn argue that for the humanities to proactively contribute to open knowledge at the global scale, new ways of thinking are needed within every part of the system. In the open information economy, the humanities are on a trajectory following the sciences, but parts of the world are almost completely left out. A cultural shift is required for universities to unlock the powerful potential of humanities open scholarship. In this wide-ranging overview, the authors show why and how the global research community must work together for meaningful outcomes. Open scholarship has undergone a profound change since its beginnings from a call to action to an essential principle in research organizations internationally. However, the core impulse remains: to reshape the information environment and harness the world’s knowledge for the greatest benefit of society. The eBook editions of this book are available open access under a CC BY-NC-ND 4.0 licence on bloomsburycollections.com. Open access was funded by Edith Cowan University.
Ovaj rad analizira filozofski koncept kompatibilizma Daniela Dennetta, s naglaskom na njegovim sociološkim implikacijama. Dennett zagovara ideju da ljudska sloboda može suživjeti s determinizmom, što se istražuje kroz kritički osvrt na njegovu teoriju iz sociološke perspektive. Prva kritika proizlazi iz zabrinutosti da prihvaćanje kompatibilizma može služiti kao racionalizacija za očuvanje društvenog statusa quo. Analizirajući perspektive društvenog determinizma i sociokulturne teorije kognitivnog razvoja, raspravlja se o utjecaju društvenih faktora na slobodnu volju pojedinca. Dodatno, koncept agencije iz sociološke perspektive razmatra se kroz prizmu međuodnosa pojedinaca i društva. Kroz primjere zajedničke agencije, ukazuje se na složenu povezanost između slobodne volje i društvenih uvjeta. Unatoč Dennettovim tvrdnjama, sociološki pristupi sugeriraju potrebu za sveobuhvatnim razumijevanjem formiranja ljudske autonomije, uzimajući u obzir društvene utjecaje. Ovaj rad ističe nužnost pristupa temi slobodne volje i njezine koegzistencije s determinizmom kroz sociološku prizmu.
Social Sciences, History of scholarship and learning. The humanities
Bui Quang Hung, Tran Anh Hoa, Tu Thanh Hoai
et al.
This study addresses the compelling need for emerging-market business organizations to excel in turbulent environments. Drawing upon a dynamic capability framework, we delve into organizational mindfulness, resilience, and performance dynamics within emerging-market contexts. Through a moderated mediation model, we investigate the impact of organizational mindfulness on organizational performance, which is mediated by organizational resilience. Additionally, we explore the moderating role of digitalized management accounting systems (MASs) use in enhancing the relationship between organizational mindfulness and organizational resilience. Utilizing partial least squares structural equation modeling (PLS-SEM), we analyze data from a two-phase survey involving 441 managers in Vietnamese organizations. Results confirm organizational resilience as a vital mediator between organizational mindfulness and organizational performance, and digitalized MAS use positively moderates the effect of organizational mindfulness on organizational resilience. This research adds to the literature on the interface between mindfulness and resilience and informs practical strategies for navigating uncertainty. By studying the interplay among mindfulness, resilience, and digitalization, we provide guidance for emerging-market organizations aiming to respond to environmental turbulence successfully.
History of scholarship and learning. The humanities, Social Sciences
In reinforcement learning, the objective is almost always defined as a \emph{cumulative} function over the rewards along the process. However, there are many optimal control and reinforcement learning problems in various application fields, especially in communications and networking, where the objectives are not naturally expressed as summations of the rewards. In this paper, we recognize the prevalence of non-cumulative objectives in various problems, and propose a modification to existing algorithms for optimizing such objectives. Specifically, we dive into the fundamental building block for many optimal control and reinforcement learning algorithms: the Bellman optimality equation. To optimize a non-cumulative objective, we replace the original summation operation in the Bellman update rule with a generalized operation corresponding to the objective. Furthermore, we provide sufficient conditions on the form of the generalized operation as well as assumptions on the Markov decision process under which the globally optimal convergence of the generalized Bellman updates can be guaranteed. We demonstrate the idea experimentally with the bottleneck objective, i.e., the objectives determined by the minimum reward along the process, on classical optimal control and reinforcement learning tasks, as well as on two network routing problems on maximizing the flow rates.
In this paper, we address the issue of fairness in preference-based reinforcement learning (PbRL) in the presence of multiple objectives. The main objective is to design control policies that can optimize multiple objectives while treating each objective fairly. Toward this objective, we design a new fairness-induced preference-based reinforcement learning or FPbRL. The main idea of FPbRL is to learn vector reward functions associated with multiple objectives via new welfare-based preferences rather than reward-based preference in PbRL, coupled with policy learning via maximizing a generalized Gini welfare function. Finally, we provide experiment studies on three different environments to show that the proposed FPbRL approach can achieve both efficiency and equity for learning effective and fair policies.
Learning rate schedules used in practice bear little resemblance to those recommended by theory. We close much of this theory/practice gap, and as a consequence are able to derive new problem-adaptive learning rate schedules. Our main technical contribution is a refined analysis of learning rate schedules for a wide class of optimization algorithms (including SGD). When considering only worst-case analysis, our theory predicts that the optimal choice is the linear decay schedule where the step-size is set proportional to 1 - t/T, where t is the current iteration and T is the total number of steps. To go beyond this worst-case analysis, we use the observed gradient norms to derive schedules refined for any particular task. These refined schedules exhibit learning rate warm-up and rapid learning rate annealing near the end of training. Ours is the first systematic approach to automatically yield both of these properties. We perform the most comprehensive evaluation of learning rate schedules to date, evaluating across 10 diverse deep learning problems, a series of LLMs, and a suite of logistic regression problems. We validate that overall, the linear-decay schedule outperforms all commonly used default schedules including cosine annealing. Our adaptive schedule refinement method gives further improvements.
The aim of this study is to analyze and examine the current situation and shortcomings of the application of experiential learning in high school history teaching and the possible improvement measures. In fact, promoting the implementation of experiential learning in high school history teaching is of great significance in eliminating students' unfamiliarity with history, developing historical skills, and improving the practicality of history learning. Despite the fact that experiential learning has been the subject of extensive research, it has been rarely applied in the traditional course syllabus, especially in the classroom teaching of the subject. This paper provides an in-depth look at how experiential learning can be further utilized effectively in the high school history classroom from both the teacher and student perspectives. The introduction part explains the background of research related to experiential learning and the significance of the research. In the first main body part, I explore the definition of experiential learning in high schools, the current status of its application, and the specific disciplines in which it is applied. The second main body part analyzes the shortcomings of experiential learning in its current application. The third part then analyzes possible solutions to the above flaws and dilemmas. Overall, this study will inform how experiential learning can be further applied in the 21st-century secondary history classroom and provide teachers with a reference for evaluating experiential learning. Each section is explained and accordingly, possible improvements are listed based on the effectiveness of the measures discussed.
While the literature develops an adequate understanding of various socio-cultural and psychological factors that contribute to expatriates’ adjustment, limited work exists regarding interdependencies between these two dimensions of adjustment, particularly in the Asian context. Using the theoretical lens of the anxiety uncertainty management (AUM) model, this research examined the various socio-cultural and psychological factors that affect expatriates’ adjustment, and the integration of these factors during adjustment. Data were collected through semi-structured interviews of 55 Chinese top management expatriates working in Pakistan, and were analyzed using thematic analysis in NVivo. Based on the findings, an AUM model of expatriates’ adjustment is proposed, which illustrates how top management expatriates may go through a four-stage process of adjustment. The various socio-cultural and psychological factors, some identified from the literature and others through fieldwork, that play a role in expatriates’ adjustment at each stage, and the integration of these factors, is discussed.
History of scholarship and learning. The humanities, Social Sciences
Several factors came together in the formation of the political, military and commercial importance of the city of Ashkelon, as it is considered the northern gateway for Egypt to penetrate the Levant and control its cities, and its commercial importance is no less important than its political position, as it is an important commercial port, as it passes through trade between the East And the West, so we find Egyptian politics always keen to have Ashkelon under its protections and subordination, and this situation exists as long as the Egyptian authority is strong, except that in the last periods of BC and after the weakness of the Egyptian authority it came under the Assyrian rule and the gate from which the armies entered to invade Egypt, These facts are all confirmed by the historical blogs on the topic.
History of scholarship and learning. The humanities
En este artículo se presentan los resultados del impacto del Pole Dance como herramienta terapéutica que favorece el empoderamiento de las mujeres. Se realizaron entrevistas a nivel internacional, dirigidas a profesionales que instruyen dicha disciplina; a las alumnas que lo practican, se les aplicó el cuestionario para medir el empoderamiento de la mujer (Hernández y García, 2008), dicha herramienta, explora 7 factores: temeridad, influencias externas, independencia, igualdad, participación, satisfacción social y seguridad, aplicándolo y analizándolo bajo el Modelo de Intervención Psicoterapéutica Centrado en Soluciones, el cual se encuentra dentro de la teoría del enfoque psicoterapéutico de la Terapia Familiar Sistémica. La finalidad de su análisis está en conocer el nivel de empoderamiento que se manifiesta en practicantes de Pole Dance, tomando en cuenta así los beneficios y las implicaciones psicológicas que derivan de dicha práctica. Encontrando que el sistema que se desarrolla en el grupo que toma clase, impacta a nivel individual, mostrando cambios de conducta y autoestima, que a su vez influyen en su sistema familiar y de pareja en los que se desenvuelve.
History of scholarship and learning. The humanities, Social Sciences
Md. Raquibuzzaman Khan, Nazia Tabassum, Niaz Ahmed Khan
et al.
Abstract The purpose of this paper is to identify and evaluate the key challenges to project procurement in public-sector agricultural development projects in Bangladesh. Being exploratory in nature, the study applied the modified Delphi method, the best worst method (BWM), and the interpretive structural modelling (ISM) approach sequentially for the investigation. Ten key procurement challenges were identified and validated through the use of a literature review and two rounds of modified Delphi with the input of 15 experts in the field. Then the BWM was applied to assess the responses of eight industry experts to estimate the relative importance of the challenges. After that, a second panel of ten experts was interviewed using ISM to look at the contextual relationships between the challenges. This led to a four-layer interpretive structural model and MICMAC (cross-impact matrix multiplication applied to classification) analysis of the challenges. Among the 10 key challenges, ‘lack of competent procurement staff’ is found to be the most significant challenge; whereas, based on the inter-relationships among the challenges, ‘political influence’ is identified as the most influential challenge. As a result, it is recommended that relevant professionals and policymakers address these challenges in terms of their relevance, relative dependencies, and influences in a holistic manner. This study addresses a knowledge gap by offering a thorough investigation of the challenges associated with public-sector agricultural project procurement in a developing country’s context. This makes it useful for professionals in the field, academics, policymakers, and future researchers.
History of scholarship and learning. The humanities, Social Sciences
Mohamad Mohamad, Julian Neubert, Juan Segundo Argayo
Federated Learning is a recent approach to train statistical models on distributed datasets without violating privacy constraints. The data locality principle is preserved by sharing the model instead of the data between clients and the server. This brings many advantages but also poses new challenges. In this report, we explore this new research area and perform several experiments to deepen our understanding of what these challenges are and how different problem settings affect the performance of the final model. Finally, we present a novel approach to one of these challenges and compare it to other methods found in literature.
Hannes Stärk, Octavian-Eugen Ganea, Lagnajit Pattanaik
et al.
Predicting how a drug-like molecule binds to a specific protein target is a core problem in drug discovery. An extremely fast computational binding method would enable key applications such as fast virtual screening or drug engineering. Existing methods are computationally expensive as they rely on heavy candidate sampling coupled with scoring, ranking, and fine-tuning steps. We challenge this paradigm with EquiBind, an SE(3)-equivariant geometric deep learning model performing direct-shot prediction of both i) the receptor binding location (blind docking) and ii) the ligand's bound pose and orientation. EquiBind achieves significant speed-ups and better quality compared to traditional and recent baselines. Further, we show extra improvements when coupling it with existing fine-tuning techniques at the cost of increased running time. Finally, we propose a novel and fast fine-tuning model that adjusts torsion angles of a ligand's rotatable bonds based on closed-form global minima of the von Mises angular distance to a given input atomic point cloud, avoiding previous expensive differential evolution strategies for energy minimization.