Evolutionary Computing

This paper presents a system that implements a society of agent performers that evolve expressive music performances (EMP) through their interactions. Each agent performer evaluates a performance using a fitness function derived from the structure of the performing piece. Agents are born with different fitness functions, representing preferences for different types of performance. A new-born agent firstly applies a Genetic Algorithm to generate a refined performance with its fitness function. Through an agent's life, it listens to the performances of others' that it connects to in the social network. Given specific conditions, it may try imitate some performances and modify its own preference. An agent performer dies of age or inactivity. New agent performers are occasionally created. With this model, we aim to generate expressive performances through simulating the effect of social pressure and culture transmission.

Resumo Computação Evolutiva é um termo general usado pra fazer referencia a uma solução dos problemas computacionais planeados e com implementação em modelos de processos evolutivos Muitos dos algoritmos evolutivos propõem paradigmas biológicos. ...

Triangular Urchin, by Chaps (www.cetoine.com). Chaps has obtained an MSc in Physics at the Swiss Federal Institute of Technology. He is the developer of the ArtiE-Fract software that was used to create Triangular Urchin. Triangular Urchin (an Iterated Functions System of 2 polar functions) emerged from an urchin structure after a few generations usingArtiE-Fract. The evolutionary process was only based on soft mutations, some of them directly induced by the author.

In this paper we describe our ongoing work on the control of a tendon driven robotic hand by an adaptive learning mechanism evolved using a simulator developed over the last years. The proposed neural network allows the robotic hand to explore its own movement possibilities to interact with objects of different shape, size and material and learn how to grasp them. As the evolved neural controller is highly adaptive, it will allow us in the future to systematically investigate the interplay between morphology and behavior using the same, but adaptive neural controller.

The case of the Zodiac Killer is one of the most widely known unsolved serial killer cases in history. The unidentified killer murdered five known victims and terrorized the state of California. He also communicated extensively with the press and law enforcement. Besides his murders, Zodiac was known for his use of ciphers. The first Zodiac cipher was solved within a week of its publication, while the second cipher was solved by the authors after 51 years, when it was discovered to be a transposition and homophonic substitution cipher with unusual qualities. In this paper, we detail the historical significance of this cipher and the numerous efforts which culminated in its solution.

We present compacton-like solution of the modified KdV equation and compare its properties with those of the compactons and solitons. We further show that, the nonlinear Schrödinger equation with a source term and other higher order KdV-like equations also possess compact solutions of the similar form.

Compactons are new class of localized solutions for families of fully nonlinear, dispersive, partial differential equa-tions. Unlike the solitons, which, although highly localized, still have infinite span, these solutions have compact support; they vanish identically outside a finite ...

It is necessary to determine the robot's pose to localize a mobile robot in a known environment. If there is no information about the initial location, we are talking about global localization. In our previous work we solved this problem in a two-dimensional environment with an algorithm known as Evolutive Localization Filter (ELF). Based on evolutionary computation concepts, the proposed algorithm searches stochastically along the state space for the robot's best pose estimate. In this paper we have improved our method designing an algorithm with new features, called ELF-3D, that uses three-dimensional sensorial information. The resulting global localization module has been tested in a simulated indoor environment. The most important improvement obtained is the accuracy of the method, allowing to apply it in manipulation tasks.

This work aims to apply parametric design in order to minimize the embodied greenhouse gas emissions and operational energy in a zero emission building in Oslo, Norway. An original generative workflow based on parametric design was developed in the Grasshopper environment to conduct energy analyses such as solar radiation and daylighting, and environmental impact analysis, in order to evaluate the embodied and operational greenhouse gas emissions of the building. The workflow was generated in order to parametrically control several building features while varying the building shape, the dimensions of construction components and the quantity of materials. The process leads to the generation of shapes with the least environmental impact. The workflow allows the modification of the initial shape of the Base Case by running iterative simulations through the Galapagos and Octopus evolutionary solvers. For each stage of the shape's optimization, through passive and active strategies, the embodied emissions and energy balances were estimated in order to evaluate how the building design would vary in terms of energy and environmental impact and to identify the implications for the design. This paper shows how design options with low levels of embodied emissions can be generated and optimized automatically, and also demonstrates how a parametric design approach provides the designer with suggestions of low-impact solutions, which can then be integrated and considered early in, and throughout, the design process in a holistic manner.

Data mining is the process of extracting desirable knowledge or interesting patterns from existing databases for specific purposes. Most conventional data-mining algorithms identify the relationships among transactions using binary values. Transactions with quantitative values are however commonly seen in real-world applications. The fuzzy concepts are used to represent item importance, item quantities, minimum supports and minimum confidences. Each attribute uses only the linguistic term with the maximum cardinality in the mining process. The number of items is thus the same as that of the original attributes, making the processing time reduced. A fuzzy-genetic data-mining algorithm for extracting both association rules and membership functions from quantitative transactions is shown in this paper. It used a combination of large 1-itemsets and membership-function suitability to evaluate the fitness values of chromosomes. The calculation for large 1-itemsets could take a lot of time...

The short form of Very Highly Advanced Artificial Intelligence is VHAAI. The short form of International Society for VHAAI is ISVHAAI. VHAAI field is used by ISVHAAI Artificial Intelligence Society to address various problems. GOD Devotee Human Swarm Optimization (GDHSO) is the novel algorithm designed in this Letter No. 13 of ISVHAAI AI Society Letters.

Instance-based methods of classification are easy to implement, easy to explain and relatively robust. Furthermore, they have often been found in empirical studies to be competitive in accuracy with more sophisticated classification techniques (

The growing sophistication of cyber threats has heightened the need for enhanced security measures in distributed web systems. This paper explores the potential of genetic algorithms to reinforce these systems by improving their adaptability to emerging vulnerabilities. Genetic algorithms, rooted in evolutionary principles, provide a mechanism to optimize security parameters, such as threat detection, intrusion response, and anomaly identification. Unlike traditional methods, these algorithms offer dynamic solutions capable of evolving with changing attack patterns. The study delves into various ways genetic algorithms can be incorporated into security frameworks for distributed systems, focusing on how these adaptive strategies could increase system resilience and robustness. Key considerations include the efficiency of configuration optimization, the ability to manage diverse security scenarios, and the scalability of algorithmic integration across multiple nodes within a system. By analyzing theoretical models and current advancements in evolutionary computing, we provide a comprehensive evaluation of the algorithm's potential impact on distributed security infrastructures. This research highlights the importance of innovative, adaptive approaches to maintaining the security of increasingly interconnected web systems in the face of evolving cyber risks.

Abstract. We dwell in largely non-technical terms on the essential differences between single-objective optimization and multiple-objective optimization. We argue in particular that single-objective approaches to real-world problems are almost invariably simplifications of the real-problem which make many ideal solutions unreachable to the optimization method. We promote the use of multi-objective optimization methods, particularly those arising from the evolutionary computation community. We point out that the state of the art in the field of ...

We conduct laboratory experiments for the multi-unit Vickrey auction with and without advice to subjects on strategy-proofness. The rate of truth-telling among the subjects without advice stays at 20%, whereas the rate increases to 47% among those who have received advice. By conducting similar experiments for the pay-your-bid auction, which is not strategy-proof, we confirm that the increase in truth-telling is due significantly to the net advice effect (i.e., the effect beyond the so-called experimenter demand effect). Moreover, we find that providing advice improves efficiency in the Vickrey auction, particularly in the early periods, when the subjects are less experienced. In general, subjects tend to overbid in Vickrey auction experiments. Our results indicate the possibility that providing simple advice decreases such overbidding by promoting a better understanding of the strategy-proofness of the Vickrey auction. Strategy-proof mechanisms are sometimes criticized because players often fail to recognize the benefit of telling the truth. However, our observations show that introducing advice on the property of strategy-proofness helps them behave "correctly.

There has been much success in analytic and computational neuroscience in establishing dynamic models of neuron activity. Lapique, Hodgkin Huxley, Fitzhugh Nagumo, Morris Lecar, Hindmarsh Rose amongst the most well known. Whilst it can be reasonably claimed that the neuron is the "atom", the basic building block of the brain, it is equally clear that the neuron itself does not capture the dynamics of segments of the brain, let alone the brain itself. For that one needs to consider large complex networks of neurons where each node (neuron) has its own dynamic. Good examples of effective models in this space are to be found in the synchronization of neurons (and other oscillatory nodes) across networks and the diffusion of signals. What is proposed here is an indirect approach based upon brain waves. Neuron networks in the brain will emit signals as a consequence of spiking activity. Therefore there will be a nonlinear time series associated with a given segment of an active brain. These series can be analyzed using techniques from nonlinear time series analysis particularly singular spectrum analysis and delayed embedding. Simulated data from a Matlab routine published in a well known paper by Izhikevich (referenced below) was subjected to these techniques and an attractor was uncovered Similarly attractors were uncovered from nonlinear time series analysis applied to simulated results from a brain simulator (Brian2) The attractors differed according the the variability of the types of neurons in the simulation. Using software to simulate networks of neurons gives a great deal of flexibility both in respect to network topology and types of behavior ascribed to neurons. The paper demonstrates the feasibility of subjecting the output from neural network simulations to SSA and Delayed Embedding to recover some interesting attractors.. The resulting constructed attractors could provide a basis for comparison with attractors recovered from real brain data.

Evolutionary systems have been used in a variety of applications, from turbine design to scheduling problems. The basic algorithms are similar in all these applications, but the representation is always problem specific. Unfortunately, the search time for evolutionary systems very much depends on efficient codings, using problem specific domain knowledge to reduce the size of the search space. This paper describes an approach, where the user only specifies a very general, basic coding that can be used in a larger variety of problems. The system then learns a more efficient, problem specific coding. To do this, an evolutionary system with variable length coding is used. While the system optimizes an example problem, a meta process identifies successful combinations of genes in the population and combines them into higher level evolved genes. The extraction is repeated iteratively, allowing genes to evolve that have a high level complexity and encode a high number of the original, basic genes. This results in a continuous restructuring of the search space, allowing potentially successful solutions to be found in much shorter search time. The evolved coding can then be used to solve other, related problems. While not excluding any potentially desirable solutions, the evolved coding makes knowledge from the example problem available for the new problem. The paper shows an example from the domain of two-dimensional shape designs. In this example, a coding is evolved that uses 366 evolved gene, encoding on average 17 basic genes. The evolved coding is successfully used to optimize a room layout, where the original coding converges to a local maximum in fitness and therefore does not find any feasible solutions.

Mohammad Khoshnevisan received his PhD degree specialising in computational finance from the University of Melbourne, Australia. In his doctoral thesis he developed an optimal control model for borrowing in the Australian petroleum industry. He has published a vast number of articles in international journals, conference proceedings and refereed monographs and many of his works are cited by elite academic/research organisations such as the American Mathematical Society and the International Statistical Institute. He

An evolutionary computation approach is described for the classical geophysics inverse problem of magnetotelluric inversion. This kind of solution to the problem is formulated as a stochastic, iterative optimisation problem, where the evolutionary algorithm operates as the optimiser. Some aspects of the approach are described, in particular three problemspecific operators that were defined: local search, homogenisation and spatial crossover. Comparison is then made between the solution obtained with the evolutionary computation method and another, that relies on a deterministic optimiser plus an entropy-based regularisation. Our results suggest that the evolutionary solution is more robust than the other, more classical approach.

The sequence design is a crucial problem in DNA based computation. We present results about tests carried on sets of DNA strands proposed in DNA computing papers since 1994. Our goal is to direct attention on the necessity of a middle-process between logical designing of input and practical computation in laboratory. The analysis shows that many input sets should not be used for a real DNA computation because they lead to a high probability of incurring in biological faults which result to be very unsafe for a reliable computation.

Fuzzy systems comprise one of the models best suited to function approximation problems, but due to the non linear dependencies between the parameters that define the system rules, the solution search space for this type of problems contains many local optima. Another important issue is the identification of the optimum structure for the fuzzy system. Depending on the complexity of the model, different solutions can be found with different compromises between their approximation error and their generalization properties. Thus, the problem becomes a multi-objective problem with two clearly competing objectives, the complexity of the model and its approximation error. The algorithms proposed in the literature to construct fuzzy systems from examples usually refine iteratively a unique model until a compromise between its complexity and its approximation error is found. This is not an adequate approach for this problem because there exists a set of Pareto-optimum solutions that can be considered equivalent. Thus, we propose the use of multi-objective evolutionary algorithms because, as they maintain a population of potential solutions for the problem, they are able to optimize both objectives simultaneously. We also incorporate some new expert evolutionary operators that try to avoid the generation of worse solutions in order to accelerate the convergence of the algorithm. The proposed algorithm is tested with some target functions widely used in the literature and the results obtained are compared to other approaches.

We consider the problem of estimating the geographic locations of nodes in a wireless sensor network where most sensors are without an effective self-positioning functionality. We propose LSVM -a novel solution with the following merits. First, LSVM localizes the network based on mere connectivity information (i.e., hop counts only), and, therefore, is simple and does not require specialized ranging hardware or assisting mobile devices as in most existing techniques. Second, LSVM is based on Support Vector Machine (SVM) learning. Although SVM is a classification method, we show its applicability to the localization problem and prove that the localization error can be upper-bounded by any small threshold given an appropriate training data size. Third, LSVM addresses the border and coverage-hole problems effectively. Last but not least, LSVM offers fast localization in a distributed manner with efficient use of processing and communication resources. We also propose a modified version of mass-spring optimization to further improve the location estimation in LSVM. The promising performance of LSVM is exhibited by our simulation study.

Aiming at investigating the roles of rotation and magnetic fields on AGB stars, the rotating version of the ATON stellar evolution code is being extended in order to account for intermediate--mass stars and their later evolutionary stages. Here we report some preliminary results on the effects of rotation and of a large-scale magnetic field on the structure and evolution of 3 and 5 M⊙ stellar models from the pre-main sequence up to the AGB.

The Rossby number Ro is an important quantity related to the well-known magnetic activity-rotation correlation for main sequence, solar-type stars. For such stars, Ro can be obtained by a semi-empirical relationship between the convective turnover time τc and the B-V colour index, but an equivalent activity-rotation correlation seems not to exist for pre-main sequence stars. In this work we report theoretical estimates of τc for low-mass, rotating pre-main sequence stars under either the Full Spectrum of Turbulence (FST) or the classical Mixing Length Theory (MLT) convection models. The results for the MLT models show that the lower the convection efficiency the higher τc, while the FST models give τc lower than those for the MLT. The presence of a parametric magnetic field lowers the convection efficiency, resulting in smaller τc values.

We report our present efforts for introducing magnetic fields in the ATON stellar evolution code code, which now evolved to truly modifying the stellar structure equations so that they can incorporate the effects of an imposed, large-scale magnetic field. Preliminary results of such an approach, as applied to low-mass stellar models, are presented and discussed.

Context. Rotational evolution in the pre-main sequence is described with new sets of pre-MS evolutionary tracks including rotation, non-gray boundary conditions (BCs) and either low (LCE) or high convection efficiency (HCE). Aims. Using observational data and our theoretical predictions, we aim at constraining (1) the differences obtained for the rotational evolution of stars within the ONC by means of these different sets of new models; (2) the initial angular momentum of low mass stars, by means of their templates in the ONC. Methods. We discuss the reliability of current stellar models for the pre-MS. While the 2D radiation hydrodynamic simulations predict HCE in pre-MS, semi-empirical calibrations either seem to require that convection is less efficient in pre-MS than in the following MS phase (lithium depletion) or are still contradictory (binary masses). We derive stellar masses and ages for the ONC by using both LCE and HCE. Results. The resulting mass distribution for the bulk of the ONC population is in the range 0.2-0.4 M for our new non-gray models and, as in previous analyse, in the range 0.1-0.3 M for models having gray BCs. In agreement with Herbst et al. ( ) we find that a large percentage (∼70%) of low-mass stars (M < ∼ 0.5 M for LCE; M < ∼ 0.35 M for HCE) in the ONC appears to be fast rotators (P < 4 days). Three possibilities are open: 1) ∼70% of the ONC low mass stars lose their disk at early evolutionary phases; 2) their "locking period" is shorter; 3) the period evolution is linked to a different morphology of the magnetic fields of the two groups of stars. We also estimate the range of initial angular momentum consistent with the observed periods. Conclusions. The comparisons made indicate that a second parameter is needed to describe convection in the pre-MS, possibly related to the structural effect of a dynamo magnetic field.

This paper proposes a new hybrid approach based on nonlinear chaotic dynamics and evolutionary strategy to forecast electricity loads and prices. The main idea is to develop a new training or identification stage in a nonlinear chaotic dynamic based predictor. In the training stage five optimal parameters for a chaotic based predictor are searched through an optimization model based on evolutionary strategy. The objective function of the optimization model is the mismatch minimization between the multistep-ahead forecasting of predictor and observed data such as it is done in identification problems. The first contribution of this paper is that the proposed approach is capable of capturing the complex dynamic of demand and price time series considered resulting in a more accuracy forecasting. The second contribution is that the proposed approach run on-line manner, i.e. the optimal set of parameters and prediction is executed automatically which can be used to prediction in real-time, it is an advantage in comparison with other models, where the choice of their input parameters are carried out off-line, following qualitative/experience-based recipes. A case study of load and price forecasting is presented using data from New England, Alberta, and Spain. A comparison with other methods such as autoregressive integrated moving average (ARIMA) and artificial neural network (ANN) is shown. The results show that the proposed approach provides a more accurate and effective forecasting than ARIMA and ANN methods.

This work analyses the effect of the Helium content on synthetic Period-Luminosity Relations (PLRs) and Period-Wesenheit Relations (PWRs) of Cepheids and the systematic uncertainties on the derived distances that a hidden population of He-enhanced Cepheids may generate. We use new stellar and pulsation models to build a homogeneous and consistent framework to derive the Cepheid features. The Cepheid populations expected in synthetic colour-magnitude diagrams of young stellar systems (from 20 Myr to 250 Myr) are computed in several photometric bands for Y = 0.25 and Y = 0.35, at a fixed metallicity (Z = 0.008). The PLRs appear to be very similar in the two cases, with negligible effects (few %) on distances, while PWRs differ somewhat, with systematic uncertainties in deriving distances as high as ∼ 7% at log P < 1.5. Statistical effects due to the number of variables used to determine the relations contribute to a distance systematic error of the order of few percent, with values decreasing from optical to near-infrared bands. The empirical PWRs derived from multi-wavelength datasets for the Large Magellanic Cloud (LMC) is in a very good agreement with our theoretical PWRs obtained with a standard He content, supporting the evidence that LMC Cepheids do not show any He effect.

As the information proliferates, searching for relevant information has become a primary task. Searching or Information retrieval (IR) aims to help the users in organising as well as retrieving those documents from the documentary collection which are most likely to satisfy information needs of the user. An optimal Information Retrieval System (IRS) is one which retrieves only those documents from the document database which are pertinent to user's information needs, while excluding documents that are not relevant. Genetic Algorithm is described by higher likelihood of finding good solutions to large and complex problems of IR optimisation. The performance of Genetic Algorithm depends upon the decision of underlying operators used namely selection, crossover and mutation. A GA-based algorithm IRIGA (Information Retrieval Improvement using Genetic Algorithm) is developed to improve the performance of Information Retrieval System. This paper presents a comparison of performance of IRIGA when different selection methods are used. The results are analysed by conducting experiments keeping the rest of the GA parameters as constant and varying only the selection strategy.

When attempting to optimize the design of engineered systems, the analyst is frequently faced with the demand of achieving several targets (e.g. low costs, high revenues, high reliability, low accident risks), some of which may very well be in con¯ict. At the same time, several requirements (e.g. maximum allowable weight, volume etc.) should also be satis®ed. This kind of problem is usually tackled by focusing the optimization on a single objective which may be a weighed combination of some of the targets of the design problem and imposing some constraints to satisfy the other targets and requirements. This approach, however, introduces a strong arbitrariness in the de®nition of the weights and constraints levels and a criticizable homogenization of physically different targets, usually all translated in monetary terms. The purpose of this paper is to present an approach to optimization in which every target is considered as a separate objective to be optimized. For an ef®cient search through the solution space we use a multiobjective genetic algorithm which allows us to identify a set of Pareto optimal solutions providing the decision maker with the complete spectrum of optimal solutions with respect to the various targets. Based on this information, the decision maker can select the best compromise among these objectives, without a priori introducing arbitrary weights.

The "ZIP" adaptive automated trading algorithm has been demonstrated to outperform human traders in experimental studies of continuous double auction (CDA) markets populated by mixtures of human and "software robot" traders. Previous papers have shown that values of the eight parameters governing behavior of ZIP traders can be automatically optimized using a genetic algorithm (GA), and that markets populated by GA-optimized traders perform better than those populated by ZIP traders with manually-set parameter values. This paper introduces a more sophisticated version of the ZIP algorithm, called "ZIP60", which requires the values of 60 parameters to be set correctly. ZIP60 is shown here to produce significantly better results in comparison to the original ZIP algorithm (called "ZIP8" hereafter) when a GA is used to search the 60-dimensional parameter space. It is also demonstrated here that this works best when the GA itself has control over the dimensionality of the search-space, allowing evolution to guide the expansion of the search-space up from 8 parameters to 60 via intermediate steps. Principal component analysis of the best evolved ZIP60 parameter-sets establishes that no ZIP8 solutions are embedded in the 60-dimensional space. Moreover, some of the results and analysis presented here cast doubt on previously-published ZIP8 results concerning the evolution of new 'hybrid' auction mechanisms that appeared to be improvements on the CDA: it now seems likely that those results were actually consequences of the relative lack of sophistication in the original ZIP8 algorithm, because "hybrid" mechanisms occur much less frequently when ZIP60s are used.

algorithmic trading, online auction marketplaces, emarketplaces, automated market mechanism design, trader-agents, ZIP traders, genetic algorithms The Zero-Intelligence Plus ("ZIP") adaptive automated trading algorithm has been demonstrated to outperform human traders in experimental studies of continuous double auction (CDA) markets populated by mixtures of human and "robot" traders . To successfully populate a market with ZIP traders, the values of eight control parameters need to be set correctly. While these eight values can be set manually, previous papers have demonstrated that values of those parameters can be automatically optimized using a genetic algorithm (GA), to tailor ZIP traders to particular markets, and also (by adding an additional real-valued numeric parameter) to automatically discover novel new forms of auction market mechanism that are more efficient than the CDA. This paper introduces a more sophisticated version of the ZIP algorithm, which is shown to produce significantly better results. The extended variant is known as "ZIP60", because it requires 60 real-valued control parameters to be set correctly, and the original ZIP algorithm is re-named "ZIP8" accordingly. Manually choosing the correct values for 60 control parameters would be a very laborious task, but it is demonstrated here that an appropriate automatic optimization process can discover good sets of values for the parameters. A simple GA operating in the 60dimensional parameter space is shown to produce ZIP60 traders with mean scores significantly improved over ZIP8s, but also with high variance in those improvements. A slight revision of this approach is shown to give results with even higher mean improvements and also with lower variance in those improvements. The revised approach involves giving the GA control over the dimensionality of the parameter space being searched, starting with an eight-dimensional space and then allowing the GA to automatically and gradually expand that space up to sixtydimensional only when the increased number of parameters leads to identifiably better solutions. The results from ZIP60, while better than ZIP8, show a greatly reduced incidence of cases where the GA discovers auction mechanisms that are significantly better than the fixed CDA mechanism. This may be interpretable as evidence that the earlier ZIP8 results (where improvements on the CDA were common) were consequences of the relative lack of sophistication in the ZIP8 algorithm.

The "ZIP" adaptive trading algorithm has been demonstrated to outperform human traders in experimental studies of continuous double auction (CDA) markets. The original ZIP algorithm requires the values of eight control parameters to be set correctly. A new extension of the ZIP algorithm, called ZIP60, requires the values of 60 parameters to be set correctly. ZIP60 is shown here to produce significantly better results than the original ZIP (called "ZIP8" hereafter). A genetic algorithm (GA) is used to search the 60-dimensional ZIP60 parameter space, and it finds parameter vectors that yield ZIP60 traders with mean scores significantly better than those of ZIP8s. This paper shows that this optimizing evolutionary search works best when the GA itself controls the dimensionality of the search-space, so that the search commences in an 8-d space and thereafter the dimensionality of the searchspace is gradually increased by the GA until it is exploring a 60-d space. Furthermore, the results from ZIP60 cast some doubt on prior ZIP8 results concerning the evolution of new 'hybrid' auction mechanisms that appeared to be better than the CDA.

The "ZIP" adaptive automated trading algorithm has been demonstrated to outperform human traders in experimental studies of continuous double auction (CDA) markets populated by mixtures of human and "software robot" traders. Previous papers have shown that values of the eight parameters governing behavior of ZIP traders can be automatically optimized using a genetic algorithm (GA), and that markets populated by GA-optimized traders perform better than those populated by ZIP traders with manually-set parameter values. This paper introduces a more sophisticated version of the ZIP algorithm, called "ZIP60", which requires the values of 60 parameters to be set correctly. ZIP60 is shown here to produce significantly better results in comparison to the original ZIP algorithm (called "ZIP8" hereafter) when a GA is used to search the 60-dimensional parameter space. It is also demonstrated here that this works best when the GA itself has control over the dimensionality of the search-space, allowing evolution to guide the expansion of the search-space up from 8 parameters to 60 via intermediate steps. Principal component analysis of the best evolved ZIP60 parameter-sets establishes that no ZIP8 solutions are embedded in the 60-dimensional space. Moreover, some of the results and analysis presented here cast doubt on previously-published ZIP8 results concerning the evolution of new 'hybrid' auction mechanisms that appeared to be improvements on the CDA: it now seems likely that those results were actually consequences of the relative lack of sophistication in the original ZIP8 algorithm, because "hybrid" mechanisms occur much less frequently when ZIP60s are used.

This paper proposes a number of forward VNS and reverse VNS algorithms for job-shop scheduling problem. The forward VNS algorithms are the variable neighborhood search algorithms applied to the original problem (i.e., the problem instance with the original precedence constraints). The reverse VNS algorithms are the variable neighborhood search algorithms applied to the reversed problem (i.e., the problem instance with the reversed precedence constraints). This paper also proposes a multi-VNS algorithm which assigns an identical initial solution-representing permutation to the selected VNS algorithms, runs these VNS algorithms, and then uses the best solution among the final solutions of all selected VNS algorithms as its final result. The aim of the multi-VNS algorithm is to utilize each single initial solution-representing permutation most efficiently and thus receive its best result in return.

Due to the complexity of theoretical approaches in Evolutionary Computation (EC), research has being largely performed on experimental basis. One popular measure used by the EC community is the success rate (SR), which is used alone or as part of more complex measures such as Koza's computational effort in Genetic Programming. A common practice in EC is to report just a punctual estimation of the SR, without additional information about its associated uncertainty. We aim to motivate EC researchers to adopt more rigorous practices when working with SRs. In particular, we introduce the importance of correctly reporting this measure and highlight its binomial nature. Unfortunately, this fact is usually overlooked in the literature. Considering the binomiality of the SR opens the whole corpus of binomial statistics to EA research and practice. In particular, we focus on studying several methods to compute SR confidence intervals, the factors that determine their quality in terms of coverage probability and interval length. Due to its practical interest, we also briefly discuss the number of required runs to build confidence intervals with a certain quality, providing a sound method to set the number of runs, one of the most important experimental settings in EC. Evidence suggests that Wilson is, on average, a reliable and simple method to bound an estimation of SR with confidence intervals, while the standard method, which is quite popular because of its conceptual simplicity, should be avoided in any case. However, other methods can also be of interest under certain circumstances. We encourage to report the number of trials and successes, as well as the interval, to ease further comparability of the results.

The logistics sector is a very dynamic and complex domain with large impact in our day-to-day life. It is also very sensitive to the market changes, so tools that can improve their processes are needed. This paper describes a decision support tool for a Spanish Logistics Company. The tool takes the knowlege from the human drivers, i.e. the address used to deliver a package, integrates and exploits it into the decision process. However, this is precisely the data that is often wrongly introduced by the drivers. The tool analyses and detects different mistakes (i.e. misspellings) in the addresses introduced, groups by zones the packages that have to be delivered, and proposes the routes that the drivers should follow. To achieve this, the tool combines three different techniques from Artificial Intelligence. First, Data Mining techniques are used to detect and correct addresses inconsistencies. Second, Case-Based Reasoning techniques that are employed to separate and learn the most frequent areas or zones that the experienced drivers do. And finally, we use Evolutionary Computation (EC) to plan optimal routes (using the human drivers past experiences) from the learned areas, and evaluate those plans against the original route executed by the human driver. Results show that the tool can automatically correct most of the misspelling in the data.

Rhythmic activities that alternate between coherent and incoherent phases are ubiquitous in chemical, ecological, climate, or neural systems. Despite their importance, general mechanisms for their emergence are little understood. In order to fill this gap, we present a framework for describing the emergence of recurrent synchronization in complex networks with adaptive interactions. This phenomenon is manifested at the macroscopic level by temporal episodes of coherent and incoherent dynamics that alternate recurrently. At the same time, the dynamics of the individual nodes do not change qualitatively. We identify asymmetric adaptation rules and temporal separation between the adaptation and the dynamics of individual nodes as key features for the emergence of recurrent synchronization. Our results suggest that asymmetric adaptation might be a fundamental ingredient for recurrent synchronization phenomena as seen in pattern generators, e.g., in neuronal systems.

VHAAI means Very Highly Advanced Artificial Intelligence. ISVHAAI means International Society for VHAAI. VHAAI field is used by ISVHAAI Society in an attempt to solve problems. A new algorithm titled Friendship Human Swarm Optimization (FdHSO) is designed in this Letter No. 8 of ISVHAAI AI Society Letters.

DC motors are widely used in industrial applications due to their reliability and ease of control. However, precise speed regulation under varying loads and disturbances remains a challenge. This paper explores the design and performance comparison of Proportional-Integral (PI) and Proportional-Integral-Derivative (PID) controllers for DC motor speed control. The controllers were simulated in MATLAB/Simulink under set-point tracking and load disturbance scenarios. Results demonstrate that the PID controller outperforms the PI controller, achieving a 15% reduction in settling time, 20% lower overshoot, and improved disturbance rejection. Statistical analysis and graphical data validate the robustness of PID control in dynamic environments. The study highlights the PID controller's suitability for applications requiring high precision and adaptability.

Level design is an art which consists of creating the combination of challenge, competition, and interaction that players call fun and involves a careful and deliberate development of the game space. When working with procedural content generation, it is necessary to review how the game designer sets the change in difficulty throughout the different levels. In this paper we present a procedural level generator that can be used for different games and is based on a genetic algorithm. We define a fitness function that does not depend on the game or content type. This function calculates the difference between the difficulty curve defined by the designer and the difficulty curve calculated for the candidate content, so the best content is the one whose difficulty curve best fits the desired curve. To design our generator, we rely on the concept of flow, theories of fun and game design.

Estimation of distribution algorithms (EDAs) are a quite recent topic in optimization techniques. They combine two technical disciplines of soft computing methodologies: probabilistic reasoning and evolutionary computing. Several algorithms and approaches have already been proposed by di erent authors, but up to now there are very few papers showing their potential and comparing them to other evolutionary computational methods and algorithms such as genetic algorithms (GAs). This paper focuses on the problem of inexact graph matching which is NP-hard and requires techniques to ÿnd an approximate acceptable solution. This problem arises when a nonbijective correspondence is searched between two graphs. A typical instance of this problem corresponds to the case where graphs are used for structural pattern recognition in images. EDA algorithms are well suited for this type of problems. This paper proposes to use EDA algorithms as a new approach for inexact graph matching. Also, two adaptations of the EDA approach to problems with constraints are described as two techniques to control the generation of individuals, and the performance of EDAs for inexact graph matching is compared with the one of GAs.

We study the asymptotic law of a network of interacting neurons when the number of neurons becomes infinite. The dynamics of the neurons is described by a set of stochastic differential equations in discrete time. The neurons interact through the synaptic weights which are Gaussian correlated random variables. We describe the asymptotic law of the network when the number of neurons goes to infinity. Unlike previous works which made the biologically unrealistic assumption that the weights were i.i.d. random variables, we assume that they are correlated. We introduce the process-level empirical measure of the trajectories of the solutions to the equations of the finite network of neurons and the averaged law (with respect to the synaptic weights) of the trajectories of the solutions to the equations of the network of neurons. The result (theorem 3.1 below) is that the image law through the empirical measure satisfies a large deviation principle with a good rate function. We provide an analytical expression of this rate function in terms of the spectral representation of certain Gaussian processes.

Given a connected, weighted, undirected graph G = (V, E), the bi-objective Minimum Diameter-Cost Spanning Tree (bi-MDCST) Problem aims to find spanning trees on G with minimal total edge weight as well as minimal diameter. The problem is known to be NP-hard and finds application in domains ranging from transportation to network design. An exact algorithm for the problem is given in the literature, but is computationally very expensive and hence applicable to only very small graphs; some fast heuristics, a multi-objective genetic algorithm (MOGA) and a fast non-dominated sorting genetic algorithm (NSGA2) have also been proposed in the literature for solving the problem and their performance studied on larger sized full graphs containing up to 250 vertices and 31,125 edges. This paper presents a novel Hybrid Multi-objective Evolutionary Algorithm (HMOEA) for the bi-MDCST problem which uses a representation that captures the genetic information of several spanning trees of varying diameters in a single chromosome, thereby enhancing the representational power of the population. The algorithm uses heuristic-based population seeding and combines linear time recombination with a fast mutation operator to more effectively balance the exploration and exploitation of the search space. The proposed HMOEA computes the optimal Pareto-front solutions in time that is several orders of magnitude lesser than that taken by the exact method, and is shown to obtain superior performance in terms of the convergence and distribution of solutions vis-à-vis the other two extant meta-heuristics on a wide range of benchmark graph instances. Results obtained by the HMOEA are also reported for larger sized Euclidean instances than attempted hitherto in the literature, containing up to 500 vertices and 124,750 edges, and for a benchmark suite of large fully connected random edge-weight graph instances introduced in this work for more effectively comparing the performance of algorithms for the bi-MDCST problem.

Mobile computing involves bulk data transmission over the transmission media. To achieve highly reliable data transmission, wireless mobile networks require efficient reliable link connectivity, regardless of terminal mobility and, thus, a reliable traffic performance. Mobile networks consist of mobile hosts, base stations, links, etc. that are often vulnerable to failure. It is desirable to design a reliable network, in terms of services of both the base stations and the communication channels of the network, for the reliable transmission of the data. An attempt is made to employ those channels that offer a reliable communication at any given time. The objective of this study is to design an appropriate reliability-based model for channel allocation that retains the overall system reliability with acceptable system performance. The system may achieve acceptable performance not only during normal operations but also under various component failures. A genetic algorithm, which is a search procedure based on evolutionary computation, is suited to solve a class of complex optimization problems. The potential of the genetic algorithm is used, in this paper, to improve the reliability of the mobile communication system. The proposed model designs a reliable mobile communication system, irrespective of the mobile hosts that change their position due to mobility. A simulation experiment to evaluate the performance of the proposed algorithm is conducted, and results reveal the effectiveness of this model.

Mobile computing involves bulk data transmission over the transmission media. To achieve highly reliable data transmission, wireless mobile networks require efficient reliable link connectivity, regardless of terminal mobility and, thus, a reliable traffic performance. Mobile networks consist of mobile hosts, base stations, links, etc. that are often vulnerable to failure. It is desirable to design a reliable network, in terms of services of both the base stations and the communication channels of the network, for the reliable transmission of the data. An attempt is made to employ those channels that offer a reliable communication at any given time. The objective of this study is to design an appropriate reliability-based model for channel allocation that retains the overall system reliability with acceptable system performance. The system may achieve acceptable performance not only during normal operations but also under various component failures. A genetic algorithm, which is a search procedure based on evolutionary computation, is suited to solve a class of complex optimization problems. The potential of the genetic algorithm is used, in this paper, to improve the reliability of the mobile communication system. The proposed model designs a reliable mobile communication system, irrespective of the mobile hosts that change their position due to mobility. A simulation experiment to evaluate the performance of the proposed algorithm is conducted, and results reveal the effectiveness of this model.

Retrieval systems accumulate a great range of documents, from abstracts, newspaper articles, and web pages to journal articles, books, court transcripts, and legislation. Collections of different types of documents represent shortcomings in current approaches toward ranking schemes. The use of short fragments of documents, called passages, instead of whole documents can overcome these shortcomings. Passage ranking provides suitable units of text, to be returned to the user, can avoid the difficulties of comparing documents of different lengths, and enables identification of short blocks of relevant material amongst otherwise irrelevant text. This paper proposes a new method of passage retrieval called fixed length arbitrary passage retrieval for Arabic documents. This method has been discussed, implemented, and evaluated. The experiment results show that ranking with fixed arbitrary passage gives substantial improvements in retrieval effectiveness over traditional document ranking schemes, particularly for queries on collections of long documents. Ranking with arbitrary passages shows consistent improvements compared to ranking with whole documents, and to ranking with other passage types.