A New Gradient Descent Approach for Local Learning of Fuzzy Neural Models

The majority of reported learning methods for Takagi-Sugeno-Kang fuzzy neural models to date mainly focus on the improvement of their accuracy. However, one of the key design requirements in building an interpretable fuzzy model is that each obtained rule consequent must match well with the system local behaviour when all the rules are aggregated to produce the overall system output. This is one of the distinctive characteristics from black-box models such as neural networks. Therefore, how to find a desirable set of fuzzy partitions and, hence, to identify the corresponding consequent models which can be directly explained in terms of system behaviour presents a critical step in fuzzy neural modelling. In this paper, a new learning approach considering both nonlinear parameters in the rule premises and linear parameters in the rule consequents is proposed. Unlike the conventional two-stage optimization procedure widely practised in the field where the two sets of parameters are optimized separately, the consequent parameters are transformed into a dependent set on the premise parameters, thereby enabling the introduction of a new integrated gradient descent learning approach. A new Jacobian matrix is thus proposed and efficiently computed to achieve a more accurate approximation of the cost function by using the second-order Levenberg-Marquardt optimization method. Several other interpretability issues about the fuzzy neural model are also discussed and integrated into this new learning approach. Numerical examples are presented to illustrate the resultant structure of the fuzzy neural models and the effectiveness of the proposed new algorithm, and compared with the results from some well-known methods.

The MNC as an externally embedded organization: An investigation of embeddedness overlap in local subsidiary networks

MNCs have been conceptualized as differentiated networks that, in turn, are embedded in external networks. Previous research has predominantly focused on the embeddedness of established subsidiaries into their local environment, omitting to shed light on the phenomenon of headquarters linkages to the local context which creates embeddedness overlap. We develop a model of why MNCs develop overlapping linkages to local subsidiary networks even if the subsidiaries have grown out of the initial start-up phase. Using detailed information on 168 European subsidiaries, we find that MNCs build and maintain more overlapping network ties when subsidiaries are high performers, hold important resources, operate in turbulent environments, and are closely connected to multinational actors as opposed to purely domestic firms.

Down with MNE-centric theories! Market entry and expansion as the bundling of MNE and local assets

Both Anderson and Gatignon and the Uppsala internationalization model see the initial mode of foreign market entry and subsequent modes of operation as unilaterally determined by multinational enterprises (MNEs) arbitraging control and risk and increasing their commitment as they gain experience in the target market. OLI and internalization models do recognize that foreign market entry requires the bundling of MNE and complementary local assets, which they call location or country-specific advantages, but implicitly assume that those assets are freely accessible to MNEs. In contrast to both of these MNE-centric views, I explicitly consider the transactional characteristics of complementary local assets and model foreign market entry as the optimal assignment of equity between their owners and MNEs. By looking at the relative efficiency of the different markets in which MNE and complementary local assets are traded, and at how these two categories of assets match, I am able to predict whether equity will be held by MNEs or by local firms, or shared between them, and whether MNEs will enter through greenfields, brownfields, or acquisitions. The bundling model I propose has interesting implications for the evolution of the MNE footprint in host countries, and for the reasons behind the emergence of Dragon MNEs.

Responsible Participation, Community Engagement and Policing in Transitional Societies: Lessons from a Local Crime Survey in Northern Ireland

The new structures of policing in Northern Ireland have been internationally lauded as a success, but the manner in which police-community relations are unfolding in local settings is less clear. In this article we draw on a local crime survey conducted in a Republican area in Belfast to examine residents’ views of policing and to highlight residents’ concerns about police effectiveness in dealing with crime and disorder. Drawing on Habermas’s concept of ‘responsible participation’, we also consider the role that community organisations can play in helping overcome local scepticism and developing positive forms of engagement with the police. © 2012 The Authors