Evolutionary learning of hierarchical decision rules

This paper describes an approach based on evolutionary algorithms, hierarchical decision rules (HIDER), for learning rules in continuous and discrete domains. The algorithm produces a hierarchical set of rules, that is, the rules are sequentially obtained and must therefore be tried until one is found whose conditions are satisfied. Thus, the number of rules may be reduced because the rules could be inside of one another. The evolutionary algorithm uses both real and binary coding for the individuals of the population. We tested our system on real data from the UCI repository, and the results of a ten-fold cross-validation are compared to C4.5s, C4.5Rules, See5s, and See5Rules. The experiments show that HIDER works well in practice.     

Self-adaptive systems: A survey of current approaches,research challenges and applications

Self-adaptive software is capable of evaluating and changing its own behavior, whenever the evaluation shows that the software is not accomplishing what it was intended to do, or when better functionality or performance may be possible. The topic of system adaptivity has been widely studied since the mid-60s and, over the past decade, several application areas and technologies relating to self-adaptivity have assumed greater importance. In all these initiatives, software has become the common element that introduces self-adaptability. Thus, the investigation of systematic software engineering approaches is necessary, in order to develop self-adaptive systems that may ideally be applied across multiple domains. The main goal of this study is to review recent progress on self-adaptivity from the standpoint of computer sciences and cybernetics, based on the analysis of state-of-the-art approaches reported in the literature. This review provides an over-arching, integrated view of computer science and software engineering foundations. Moreover, various methods and techniques currently applied in the design of self-adaptive systems are analyzed, as well as some European research initiatives and projects. Finally, the main bottlenecks for the effective application of self-adaptive technology, as well as a set of key research issues on this topic, are precisely identified, in order to overcome current constraints on the effective application of self-adaptivity in its emerging areas of application.     

Combined myofibrillar and mitochondrial creatine kinase deficiency impairs mouse diaphragm isotonic function

Creatine kinase (CK) is an enzyme central to cellular high-energy phosphate metabolism in muscle. To characterize the physiological role of CK in respiratory muscle during dynamic contractions, we compared the force-velocity relationships, power, and work output characteristics of the diaphragm (Dia) from mice with combined myofibrillar and sarcomeric mitochondrial CK deficiency (CK[-/-]) with CK-sufficient controls (Ctl). Maximum velocity of shortening was significantly lower in CK[-/-] Dia (14.1 +/- 0.9 Lo/s, where Lo is optimal fiber length) compared with Ctl Dia (17.5 +/- 1.1 Lo/s) (P < 0.01). Maximum power was obtained at 0.4-0.5 tetanic force in both groups; absolute maximum power (2,293 +/- 138 W/m2) and work (201 +/- 9 J/m2) were lower in CK[-/-] Dia compared with Ctl Dia (2,744 +/- 146 W/m2 and 284 +/- 26 J/m2, respectively) (P < 0.05). The ability of CK[-/-] Dia to sustain shortening during repetitive isotonic activation (75 Hz, 330-ms duration repeated each second at 0.4 tetanic force load) was markedly impaired, with CK[-/-] Dia power and work declining to zero by 37 +/- 4 s, compared with 61 +/- 5 s in Ctl Dia. We conclude that combined myofibrillar and sarcomeric mitochondrial CK deficiency profoundly impairs Dia power and work output, underscoring the functional importance of CK during dynamic contractions in skeletal muscle.