The Agent-Rule-Class Framework for Multi-Agent Systems

Multi-agent systems have become increasingly mature, but their appearance does not make the traditional OO approach obsolete. On the contrary, OO methodologies can benefit from the principles and tools designed for agent systems. The Agent-Rule-Class (ARC) framework is proposed as an approach that builds agents upon traditional OO system components and makes use of business rules to dictate agent behaviour with the aid of OO components. By modelling agent knowledge in business rules, the proposed paradigm provides a straightforward means to develop agent-oriented systems based on the existing object-oriented systems and offers features that are otherwise difficult to achieve in the original OO systems. The main outcome of using ARC is the achievement of adaptivity. The framework is supported by a tool that ensures agents implement up-to-date requirements from business people, reflecting desired current behaviour, without the need for frequent system rebuilds. ARC is illustrated with a rail track example.

A recursive rule base adjustment algorithm for a fuzzy logic controller

This paper introduces a recursive rule base adjustment to enhance the performance of fuzzy logic controllers. Here the fuzzy controller is constructed on the basis of a decision table (DT), relying on membership functions and fuzzy rules that incorporate heuristic knowledge and operator experience. If the controller performance is not satisfactory, it has previously been suggested that the rule base be altered by combined tuning of membership functions and controller scaling factors. The alternative approach proposed here entails alteration of the fuzzy rule base. The recursive rule base adjustment algorithm proposed in this paper has the benefit that it is computationally more efficient for the generation of a DT, and advantage for online realization. Simulation results are presented to support this thesis. (c) 2005 Elsevier B.V. All rights reserved.

Rule Britannia!: British Stock Market Returns, 1825-1870

This article presents a new series of monthly equity returns for the British stock market for the period 1825-1870. In addition to calculating capital appreciation and dividend yields, the article also estimates the effect of survivorship bias on returns. Three notable findings emerge from this study. First, stock market returns in the 1825-1870 period are broadly similar for Britain and the United States, although the British market is less risky. Second, real returns in the 1825-1870 period are higher than in subsequent epochs of British history. Third, unlike the modern era, dividends are the most important component of returns.

'Rule of Law and Politics of Anti-Corruption Campaigns in a Post-Authoritarian State- The Nigerian Case'

This article evaluates the anti-corruption campaign instituted in Nigeria following on the post-authoritarian transition in the country, with specific focus on political corruption. The anti-corruption campaign is being prosecuted within a context where law is as critical a factor as politics. This article examines whether the judiciary, in view of its accountability deficit, can offer legitimacy to the campaign. How has its questionable credentials impacted on its involvement in the campaign to sanitise public life? What has been the impact of the judicial role on the rule of law? These are some of the important questions this article seeks to answer. The inquiry in this article demonstrates how the guardian institution of the rule of law faces an uphill task in the performance of that role in a post-authoritarian context.

A heterosynaptic learning rule for neural networks

In this article we intoduce a novel stochastic Hebb-like learning rule for neural networks that is neurobiologically motivated. This learning rule combines features of unsupervised (Hebbian) and supervised (reinforcement) learning and is stochastic with respect to the selection of the time points when a synapse is modified. Moreover, the learning rule does not only affect the synapse between pre- and postsynaptic neuron, which is called homosynaptic plasticity, but effects also further remote synapses of the pre-and postsynaptic neuron. This more complex form of synaptic plasticity has recently come under investigations in neurobiology and is called heterosynaptic plasticity. We demonstrate that this learning rule is useful in training neural networks by learning parity functions including the exclusive-or (XOR) mapping in a multilayer feed-forward network. We find, that our stochastic learning rule works well, even in the presence of noise. Importantly, the mean leaxning time increases with the number of patterns to be learned polynomially, indicating efficient learning.