Search

We commonly assume that data are a homogeneous set of observations when learning the structure of Bayesian networks. However, they often comprise different data sets that are related but not homogeneous because they have been collected in different ways or from different populations. In our previous work (Azzimonti, Corani and Scutari, 2021), we proposed a closed-form Bayesian Hierarchical Dirichlet score for discrete data that pools information across related data sets to learn a single encompassing network structure, while taking into account the differences in their probabilistic structures. In this paper, we provide an analogous solution for learning a Bayesian network from continuous data using mixed-effects models to pool information across the related data sets. We study its structural, parametric, predictive and classification accuracy and we show that it outperforms both conditional Gaussian Bayesian networks (that do not perform any pooling) and classical Gaussian Bayesian networks (that disregard the heterogeneous nature of the data). The improvement is marked for low sample sizes and for unbalanced data sets., Comment: 12 pages, 5 figures

This paper introduces the notion of ``Initial Alignment'' (INAL) between a neural network at initialization and a target function. It is proved that if a network and a Boolean target function do not have a noticeable INAL, then noisy gradient descent on a fully connected network with normalized i.i.d. initialization will not learn in polynomial time. Thus a certain amount of knowledge about the target (measured by the INAL) is needed in the architecture design. This also provides an answer to an open problem posed in [AS20]. The results are based on deriving lower-bounds for descent algorithms on symmetric neural networks without explicit knowledge of the target function beyond its INAL.

This study investigates the role of national and organisational culture in day-to-day activities of multinational project teams, specifically focusing on differences between Chinese and Dutch project managers. We rely on fieldwork observation and interviews with representatives from a diverse set of organizations in China and the Netherlands. Analyses focus on the impact of cultural differences on five project management processes – (1) project planning, (2) cost and quality management, (3) risk management, (4) scope management and project promises, and (5) communication. Although there are many differences observed in these five processes, research subjects report no significant impact of cross-cultural collaboration on project performance. We conclude that cross-cultural project teams can provide critical elements for an effective combination of different project management practices: people from various national and organisational cultures, enriched by different experiences and management theories, with a mix of skills. This study provides insights for those who work cross culturally (especially between western and eastern contexts) and is also a contribution to both the project management and cross-cultural management literatures.

We review and integrate a wide range of literature that has examined the strategies by which organizations navigate institutionally diverse settings and capture rents outside of the marketplace. We synthesize this body of research under the umbrella term institutional strategies, which we define as the comprehensive set of plans and actions directed at strategically leveraging and shaping the socio-political and cultural institutions within an organization’s external environment. Our review of institutional strategies is focused on emerging market contexts, settings that are characterized by weak capital market and regulatory infrastructures and fast-paced turbulent change. Under such challenging conditions, strategies aimed at shaping the institutional environment may be especially critical to an organization’s performance and long-term survival. Our review reveals that organizations engage in three specific and identifiable sets of institutional strategies, which we term: relational, infrastructure-building, and socio-cultural bridging. We conclude by highlighting fruitful avenues for cross-disciplinary dialogue in the hope of promoting future research on emerging markets and defining the next frontier of institutional theory in organizational analysis.

We examine the heterogeneous effects of globalization on the interest rate setting by microfinance institutions (MFIs) around the world. We consider MFIs as a mechanism to overcome the institutional void of credit for small entrepreneurs in developing and emerging economies. Using a large global panel of MFIs from 119 countries, we find that social globalization that embraces egalitarian institutions on average reduces MFIs’ interest rates. In contrast, economic globalization that embraces neoliberal institutions on average increases MFIs’ interest rates. Moreover, the proportions of female borrowers and of poorer borrowers negatively moderate the relationship between social globalization and MFI interest rate, and positively moderate the relationship between economic globalization and MFI interest rate. This paper contributes to understanding how globalization processes can both ameliorate and exacerbate challenges of institutional voids in emerging and developing economies.