Playing with complexity: From cellular evolutionary algorithms with coalitions to self-organizing maps

The key technology concerning the dispatch and control of the cyber-physical-social systems (CPSS) integration and group machine learning (ML) based Web-of-Cells (WoC) is systematically investigated, aiming to develop an intelligent dispatching system that has a high penetration of distributed generations (DG). Based on practical engineering demands and the weakly-centralized WoC, which is characterized by self-organized co-evolution, high independence, high-efficiency synergy, and autonomous learning, a variety of advanced theoretical tools such as complex network theory, group ML, evolutionary game theory, and CPSS-based parallel system theory have been adopted to address the following key issue: How can we achieve overall optimal dispatching and control decision-making in a class of complex systems relying on a large number of group cells with characteristics of limited information, weak controllability, small capacity, and wide distribution? Starting from this, four basic scientific issues are discussed: 1) a modeling method for a self-organized coupled network of the CPSS integration-based WoC; 2) a stability analysis and stability control system of the self-organized evolution of the WoC; 3) a highly autonomous group intelligent decision (GID) method of an independent cell; 4) multi-cell synergetic evolutionary game and GID theory. Hence, an innovative breakthrough on the intersection of complex network game theory and group ML is expected to be obtained, contributing to the emergence of group knowledge in complex circumstances and a significant improvement in the level of GID. Lastly, based on the theoretical investigation in this paper, explorations on the development of the CPSS platform for the WoC as well as its engineering practice in application are conducted. Furthermore, the in-depth development of the WoC, as well as its anticipated technical challenges, are prospected and analyzed with the hope of applying it on practical smart distribution grid demonstration projects in the future.

Fine-Grained Parallel and Distributed Spatial Stochastic Simulation of Biological Reactions

The cellular genetic algorithm (CGA) is growing which combines GAs with cellular automata. The individuals are distributed in a toridal grid or presudo landscape and their genetic operator is restricted to within neighborhood. In this paper, we present a cellular genetic algorithm with evolutional rule (CGAE) which more closely mimics the process of evolvement in ecology by cellular automata. It provides a mechanism for maintaining flexible population size and self- adaptive control on migration .We investigate the performance and behavior of the algorithm on complex problem like problems having high epistasis and multimodality. According to numerical optimization problems, CGAE is compared with other CGA. In the paper, the experiment result indicated that CGAE can improve convergent speed and maintain diversity of population. It shows that this novel algorithm has application prospects.

Evolutionary Algorithms (EAs) are stochastic search algorithms inspired by the process of neo-Darwinian evolution. The motivation for applying EAs to data mining is that they are robust, adaptive search techniques that perform a global search in the solution space. This chapter first presents a brief overview of EAs, focusing mainly on two kinds of EAs, viz. Genetic Algorithms (GAs) and Genetic Programming (GP). Then the chapter reviews the main concepts and principles used by EAs designed for solving several data mining tasks, namely: discovery of classification rules, clustering, attribute selection and attribute construction. Finally, it discusses Multi-Objective EAs, based on the concept of Pareto dominance, and their use in several data mining tasks.

The cellular genetic algorithm (CGA) combines GAs with cellular automata by spreading an evolving population across a pseudo-landscape. In this study we use insights from ecology to introduce new features, such as disasters and connectivity changes, into the algorithm. We investigate the performance and behaviour of the algorithm on standard GA hard problems. The CGA has the advantage of avoiding premature convergence and outperforms standard GAs on particular problems. A potentially important feature of the algorithm’s behaviour is that the fitness of solutions frequently improves in large jumps following disturbances (culling by patches).KeywordsGenetic AlgorithmCellular AutomatonCellular Automaton ModelParallel Genetic AlgorithmRastrigin FunctionThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Explaining cooperative behaviors in biological and social systems is a fundamental and challenging problem. Evolutionary game theory provides a powerful framework addressing this issue. Costly punishment is proven to be a promising mechanism promoting cooperation by many experimental designs in the field of evolutionary game theory. We first give an introduction to cooperation and evolutionary games, and then review experimental and theoretical works on punishment, focusing on the evolution of punishment, its positive and detrimental effects on cooperation evolving. We further explore the dynamical influence induced by punishment evasion, and the effect of institutional punishment and reward on cooperation. Finally we summarize our conclusions and briefly outline directions for future research.

The data deluge is changing the operating environment of many sensing systems from data-poor to data-rich-so data-rich that we are in jeopardy of being overwhelmed. Managing and exploiting the data deluge require a reinvention of sensor system design and signal processing theory. The potential pay-offs are huge, as the resulting sensor systems will enable radically new information technologies and powerful new tools for scientific discovery.

Event Takeover Values (ETV) measure the impact of each individual in the population dynamics of evolutionary algorithms (EA). Previous studies argue that ETV distribution of panmictic EAs fit power laws with exponent between 2.2 and 2.5 and that this property is insensitive to fitness landscapes and design choices of the EAs. One exception is cellular EAs, for which there are deviations of the power law for large values. In this paper, ETVs for structured and panmictic EAs with different population size and mutation probability on several fitness landscapes were computed. Although the ETVs distribution of pamictic EAs is heavy-tailed, the log-log plot of the complementary cumulative distributed function shows no linearity. Furthermore, Vuong's tests on the distributions generated by several instances of the problems conclude that power law models cannot be favored over log-normal models. On the other hand, the tests confirm that cEAs impose significant deviations to the distribution tail.

There have been many attempts to understand complexity and to represent it in terms of computable quantities. To date, however, these attempts have had little success. Although we find complexity in a broad range of scientific domains, precise definitions escape our grasp (Bak, 1996; Morin, 2001; Prigogine & Stengers, 1984). One of the key models in complexity science is the Cellular Automaton (CA), a class of system in which small changes in the initial conditions or in local rules can provoke unpredictable behavior (Wolfram, 1984; Wolfram, 2002; Langton, 1986; 1990). The key issue, here as in other kinds of complex system, is to discover the rules governing the emergence of complex phenomena. If such rules were known we could use them to model and predict the behavior of complex physical and biological systems. Taking it for granted that complex behavior is the result of interactions among multiple components of a larger system; we can ask a number of fundamental questions.

Nanomaterials offer promising solutions for chemotherapy challenges, addressing issues like cytotoxicity and biocompatibility. In cancer clinical protocols, biomedical imaging is vital, providing insights into tumor morphology. Luminescent nanomaterials or nanoparticles (LNPs), particularly effective for diseases like cancer, possess controllable properties like size (usually <100 nm), surface charge, and external functionalization. LNPs interact with biological systems at systemic and cellular levels. Cellular uptake is crucial, allowing selective targeting of cancer cells through overexpressed surface receptors such as transferrin receptor (TfR), G-protein coupled receptor (GPCR), folate receptor (FR), epidermal growth factor receptor (EGFR), lectins, and low-density lipoprotein receptor (LDLR). LNPs can accumulate in subcellular compartments, playing a pivotal role in drug delivery. Studies explore LNPs’ internalization into cells, investigating their potential to deliver cargoes like DNA, siRNA, miRNA, and small-molecule drugs. This review highlights the latest advancements in LNPs and their biomedical applications. Despite these promising developments, comprehensive nanotoxicological assessments are crucial for a better understanding of LNPs’ behavior in biological systems, paving the way for future clinical applications.

In this paper, we cluster and analyze data from the past embedded software development projects using self-organizing maps (SOMs)[9] that are a type of artificial neural networks that rely on unsupervised learning. The purpose of the clustering and analysis is to improve the accuracy of predicting the number of errors. A SOMproduces a low-dimensional, discretized representation of the input space of training samples; these representations are called maps. SOMs are useful for visualizing low-dimensional views of high-dimensional data, a multidimensional scaling technique. The advantages of SOMs for statistical applications are as follows: (1) data visualization, (2) information processing on association and recollection, (3) summarizing large-scale data, and (4) creating nonlinear models. To verify our approach, we perform an evaluation experiment that compares SOM classification to product type classification using Welch’s t-test for Akaike’s Information Criterion (AIC). The results indicate that the SOM classification method is more contributive than product type classification in creating estimation models, because the mean AIC of SOM classification is statistically significantly lower.KeywordsSelf-organizing mapsclusteringembedded software development

The paper describes the comparison of three evolutionary algorithms for solving combinatorial optimization problems. In particular, a generational, a steady-state and a cellular genetic algorithm were applied to the maximum cut problem, the error correcting code design problem, and the minimum tardy task problem. The results obtained in this work are better than the ones previously reported in the literature in all cases except for one problem instance. The high quality results were achieved although no problem-specific changes of the evolutionary algorithms were made other than in the fitness function. The constraints for the minimum tardy task problem were taken into account by incorporating a graded penalty term into the fitness function. The generational and steady-state algorithms yielded very good results although they sampled only a tiny fraction of the search space.

In order to overcome the drawbacks of Simple Genetic Algorithm such as cannot get the most optimal result, low convergence speed et al. Cellular Simple Genetic Algorithm-a new genetic algorithm based on Cellular Automata-is presented in this paper. Compared with the Simple Genetic Algorithm, the experiment results show the Cellular Simple Genetic Algorithm has remarkable advantages in following aspects: reducing the search-time and improving the precise of target function.

Network traffic classification has always been an important part of realizing effective network management. Due to network traffic is high-dimensional nonlinear, classical Self-Organizing Maps (SOM) has poor robustness and reliability because it adopts Euclidean distance. A network traffic classification method based on Kernel-SOM (KSOM) is proposed, which replaces Euclidean distance with non-Euclidean distance induced by kernel function, and adopts it to estimate the matching degree between the input pattern and the connection weight. Experimental results demonstrate that compared with the classical SOM and NB, KSOM achieves higher classification presicion, and has shown fascinating characteristic when being used in the classification of network traffic.

Although different kinds of evolutionary algorithms (EAs) have been designed and achieved great success on many optimization problems, they are usually limited to some small-scale problems, e.g. with less than 100 decision variables, which may be quite small comparing to the requirements of real-world applications. Therefore, scaling EAs to large size problems have attracted more and more interest. Conventional EAs mimic the seemingly random natural processes by which species evolve. These evolution processes are slow or inefficient. Now, genetic engineering has enabled man to increase both the yields and quality of some crops fast by modifying some part of their genome precisely. In this paper, inspired by the ideas of the genetic engineering, we designed a local selection operator by decomposing the high-dimensional problem into some subcomponents and assigning a local fitness function to evaluate each subcomponent. Then a new differential evolution (DE) is proposed by inserting the local selection operator into the framework of DE. Numerical experiments were carried out to evaluate the performance of the new algorithm on a large number of benchmark functions. The results show that the new algorithm is effective and efficient for high-dimensional optimization.