Search results

Compared with traditional methods relying on manual teaching or system modeling, data-driven learning methods, such as deep reinforcement learning and imitation learning, show more promising potential to cope with the challenges brought by increasingly complex tasks and environments, which have become the hot research topic in the field of robot skill learning. However, the contradiction between the difficulty of collecting robot–environment interaction data and the low data efficiency causes all these methods to face a serious data dilemma, which has become one of the key issues restricting their development. Therefore, this paper aims to comprehensively sort out and analyze the cause and solutions for the data dilemma in robot skill learning.

First, this review analyzes the causes of the data dilemma based on the classification and comparison of data-driven methods for robot skill learning; Then, the existing methods used to solve the data dilemma are introduced in detail. Finally, this review discusses the remaining open challenges and promising research topics for solving the data dilemma in the future.

This review shows that simulation–reality combination, state representation learning and knowledge sharing are crucial for overcoming the data dilemma of robot skill learning.

To the best of the authors’ knowledge, there are no surveys that systematically and comprehensively sort out and analyze the data dilemma in robot skill learning in the existing literature. It is hoped that this review can be helpful to better address the data dilemma in robot skill learning in the future.

This study attempts to answer the question: “how are the two drivers, accountability focus and organizational learning, independently and interactively associated with public agencies’ proactive policy orientation?” The first driver is the multiple accountabilities that public agencies pursue: (1) bureaucratic, (2) legal, (3) professional and (4) political. The second driver is the organizational learning activities of public agencies: (1) socialization, (2) externalization, (3) combination and (4) internalization.

For data, 800 respondents from the public agencies in South Korea were surveyed.

The analysis provided several findings: (1) the discretionary accountabilities (professional and political) have a greater positive influence on the proactive policy orientation; (2) the conventional accountabilities (legal and bureaucratic) tend to have negative impacts on the proactive policy orientation and (3) among the four types of accountability, legal accountability can be more significantly complemented by organizational learning activities, which can enable both visionary and realistic administration in a balanced manner.

This study provides a unique insight on how organizational proactivity can be ensured through the interactions of organizational accountabilities and organizational learning.

This exploratory study discusses the policy learning process of the development of disaster risk reduction (DRR) policy.

The paper discusses how DRR has and has not developed in Thailand through the two major disasters: the 2004 Indian Ocean Tsunami and the 2011 Great Flood. The information was collected by documentary analysis to gain a historical and critical understanding of the development of the system and policy of DRR in Thailand. Additionally, key stakeholders' interviews were undertaken to supplement the analysis.

The paper demonstrates that Thailand's DRR development has been “reactive” rather than “proactive”, being largely directed by global DRR actors.

Being a small-scale study, the sample size was small. The analysis and argument would be consolidated with an increase in the number of interviews.

The model can help deconstruct which dimension of the learning process a government has/has not achieved well.

The application of the “restrictive-expansive policy learning” model, which identifies different dimensions of policy learning, reveals that the Thai government's policy learning was of a mixed nature.

The current synthesis study was conducted to locate comprehensive perspectives about the transformation of higher education institutions from being the only places where formal education programs are offered into settings where lifelong learning can be integrated. This demands an inquiry through not only instance investigation but also a more comprehensive evidence upsurge which has great importance in obtaining lessons and drawing conclusions from existing facts to show how higher education institutions can be places where lifelong learning is promoted for the good of both individuals and societal advancement.

Using a meta-synthesis methodology, a comprehensive overview of the current state of knowledge in the area of higher education institutions' role in promoting lifelong learning was synthesized.

The study identified wide-ranging lifelong learning conceptualizations, potential beneficiaries, learning contents and ways of delivery that can be applied in higher education institutions. Furthermore, the practical challenges, partnership and coordination concerns and policy and reform issues towards promoting lifelong learning were addressed.

This meta-synthesis provides crucial evidence for higher education policymakers and practitioners seeking to guide the transformation of their institutions into settings where lifelong learning is integrated with other forms of educational programs, thereby optimizing individual's professional development and societal progress.

Limited by the types of sensors, the state information available for musculoskeletal robots with highly redundant, nonlinear muscles is often incomplete, which makes the control face a bottleneck problem. The aim of this paper is to design a method to improve the motion performance of musculoskeletal robots in partially observable scenarios, and to leverage the ontology knowledge to enhance the algorithm’s adaptability to musculoskeletal robots that have undergone changes.

A memory and attention-based reinforcement learning method is proposed for musculoskeletal robots with prior knowledge of muscle synergies. First, to deal with partially observed states available to musculoskeletal robots, a memory and attention-based network architecture is proposed for inferring more sufficient and intrinsic states. Second, inspired by muscle synergy hypothesis in neuroscience, prior knowledge of a musculoskeletal robot’s muscle synergies is embedded in network structure and reward shaping.

Based on systematic validation, it is found that the proposed method demonstrates superiority over the traditional twin delayed deep deterministic policy gradients (TD3) algorithm. A musculoskeletal robot with highly redundant, nonlinear muscles is adopted to implement goal-directed tasks. In the case of 21-dimensional states, the learning efficiency and accuracy are significantly improved compared with the traditional TD3 algorithm; in the case of 13-dimensional states without velocities and information from the end effector, the traditional TD3 is unable to complete the reaching tasks, while the proposed method breaks through this bottleneck problem.

In this paper, a novel memory and attention-based reinforcement learning method with prior knowledge of muscle synergies is proposed for musculoskeletal robots to deal with partially observable scenarios. Compared with the existing methods, the proposed method effectively improves the performance. Furthermore, this paper promotes the fusion of neuroscience and robotics.

Machine learning (ML), and deep learning in particular, is gaining traction across a myriad of real-life applications. Portfolio management is no exception. This paper provides a systematic literature review of deep learning applications for portfolio management. The findings are likely to be valuable for industry practitioners and researchers alike, experimenting with novel portfolio management approaches and furthering investment management practice.

This review follows the guidance and methodology of Linnenluecke et al. (2020), Massaro et al. (2016) and Fisch and Block (2018) to first identify relevant literature based on an appropriately developed search phrase, filter the resultant set of publications and present descriptive and analytical findings of the research itself and its metadata.

The authors find a strong dominance of reinforcement learning algorithms applied to the field, given their through-time portfolio management capabilities. Other well-known deep learning models, such as convolutional neural network (CNN) and recurrent neural network (RNN) and its derivatives, have shown to be well-suited for time-series forecasting. Most recently, the number of papers published in the field has been increasing, potentially driven by computational advances, hardware accessibility and data availability. The review shows several promising applications and identifies future research opportunities, including better balance on the risk-reward spectrum, novel ways to reduce data dimensionality and pre-process the inputs, stronger focus on direct weights generation, novel deep learning architectures and consistent data choices.

Several systematic reviews have been conducted with a broader focus of ML applications in finance. However, to the best of the authors’ knowledge, this is the first review to focus on deep learning architectures and their applications in the investment portfolio management problem. The review also presents a novel universal taxonomy of models used.

Educational policies, integrated practices, obliged strategies and notable benchmarks are always required by the higher educational institutions (HEIs) for operating business ventures into competent boundaries and to preside toward the overall new business density. The same are needed to be evaluated based on student's concerns for road-mapping sustainability. Accordingly, authors conducted present study to identify crucial quality characteristics (measures) under the origins of HEIs based on student's concerns using qualitative medium under Indian economy. The study is presenting critical dimensions and quality characteristics, which are seeking by the students for selecting HEIs for their studies.

Kano integrated-Grey-VIKOR approach is utilized in present study for road-mapping sustainability based on the determination of priority index and ranking. The study utilized three segments of methodology, where in the first segment, Kano technique is implicated to define priority index of quality characteristics. In the second segment, grey sets theory is implicated to capture the perceptions of the respondents. In the third segment, VIKOR technique is implicate to rank the HEIs.

The findings of the study will assist administrators in planning the prominent strategies that can embrace performance traits under HEI, which in turn will participate in growth and development of an economy. The findings have revealed “PPCS, ICMC, TSTR, PICM, AFEP, IMIS as Attractive performance characteristics,” “IEAF, OIAR, INET as One dimensional performance characteristics,” “QTCS, PORE, SIRD as Must-be performance characteristics” and “PQPE, PCTM as Indifferent performance characteristics.” Additionally, “Professional and placement characteristics of institute” is found as the most significant measure inspiring students for admiring engineering institutes. It is found that “Observance of institutional affiliation and recognition” and “Infrastructure, classroom management and control methods” are found as the second significant measures. “Patterns of question papers and evaluation medium” and “Personal characteristics of teacher and management” are found as the least competent characteristics admiring stakeholders for selecting HEI.

The present study can assist administrators in drafting refined policies and strategies for practising quality outputs by HEI. The study suggested critical quality characteristics, which in respond will aid in attracting more number of students toward educational institutes. A study under Indian context is demonstrated for presenting critical facts and attaining higher student's enrolment rates.

This study aims to focus on the legal status of the Financial Action Task Force (FATF) regulatory spread in spite of its limited membership in international law. This is conducted by examining the regime of the FATF with the normative regime of public international law and trying to identify common grounds and conflicts between the two.

This study adopted an exploratory approach involving a thorough examination and analysis of accredited text, command papers and reports, archival materials, national obligations, websites as well as other documentary evidence.

This research gives an empirical determinant of compliance behaviour in response to FATF regulatory standards and the interplay of international law.

The findings here are not exhaustive and could be approached from other perspectives. Researchers are therefore encouraged to engage by testing the findings further, as this is only a blueprint for further research.

This study provides implications for the need to open up the current membership of the FATF, as it appears discriminatory in nature and could inhibit effective compliance with its regulatory standards.

FATF regulatory standards do not just revolve around its members and rule-takers but also affect unintended and vulnerable people who were never in contemplation when these regulations were debated without a global consensus.

The main aim of this study is to advocate for a rethink of FATF’s regulatory strategy by ensuring that its operations are more inclusive, where jurisdictions can participate as members, creating a sense of belonging and commitment in the fight against money laundering.

The aim of this study is to evaluate Big Data Analytics (BDA) drivers in the context of food supply chains (FSC) for transition to a Circular Economy (CE) and Sustainable Operations Management (SOM).

Ten different BDA drivers in FSC are examined for transition to CE; these are Supply Chains (SC) Visibility, Operations Efficiency, Information Management and Technology, Collaborations between SC partners, Data-driven innovation, Demand management and Production Planning, Talent Management, Organizational Commitment, Management Team Capability and Governmental Incentive. An interpretive structural modelling (ISM) methodology is used to indicate the relationships between identified drivers to stimulate transition to CE and SOM. Drivers and pair-wise interactions between these drivers are developed by semi-structured interviews with a number of experts from industry and academia.

The results show that Information Management and Technology, Governmental Incentive and Management Team Capability drivers are classified as independent factors; Organizational Commitment and Operations Efficiency are categorized as dependent factors. SC Visibility, Data-driven innovation, Demand management and Production Planning, Talent Management and Collaborations between SC partners can be classified as linkage factors. It can be concluded that Governmental Incentive is the most fundamental driver to achieve BDA applications in FSC transition from linearity to CE and SOM. In addition, Operations Efficiency, Collaborations between SC partners and Organizational Commitment are key BDA drivers in FSC for transition to CE and SOM.

The interactions between these drivers will provide benefits to both industry and academia in prioritizing and understanding these drivers more thoroughly when implementing BDA based on a range of factors. This study will provide valuable insights. The results from this study will help in drawing up regulations to prevent food fraud, implementing laws concerning government incentives, reducing food loss and waste, increasing tracing and traceability, providing training activities to improve knowledge about BDA and focusing more on data analytics.

The main contribution of the study is to analyze BDA drivers in the context of FSC for transition to CE and SOM. This study is unique in examining these BDA drivers based on FSC. We hope to find sustainable solutions to minimize losses or other negative impacts on these SC.

This paper aims to further our understanding of policy–practice gaps in organizations from an organizational learning perspective. The authors conceptualize and analyze policy–practice gaps in terms of what they label the dual challenge of organizational learning, i.e. the organizational tasks of both adapting ongoing practices to prescribed policy demands and adapting the policy itself to the needs of practice. Specifically, the authors address how this dual challenge can be understood in terms of organizational learning and how an organization can be managed to successfully resolve the dual learning challenge and, thereby, bridge policy–practice gaps in organizations.

This paper draws on existing literature to explore the gap between policy and practice. Through a synthesis of theories and an illustrative practical example, this paper highlights key conceptual underpinnings.

In the analysis of the dual challenge of organizational learning, this study provides a conceptual framework that emphasizes the important role of tensions and contradictions between policy and practice and their role as drivers of organizational learning. To bridge policy–practice gaps in organizations, this paper proposes five key principles that aim to resolve the dual challenge and accommodate both deployment and discovery in organizations.

Because this is a conceptual study, empirical research is called for to explore further and test the findings and conclusions of the study. Several avenues of possible future research are proposed.

This paper primarily contributes by introducing and elaborating on a conceptual framework that offers novel perspectives on the dual challenges of facilitating both discovery and deployment processes within organizations.