Finding Optimal Behaviors with Genetic Algorithms

Inspired by the evolution of biological systems, genetic algorithms have been applied to generate solutions for optimization problems in a variety of scientific and engineering disciplines. For a given problem, a suitable genome representation must be defined along with a mutation operator to generate subsequent generations. Unlike natural systems, which display a variety of complex rearrangements (e.g., mobile genetic elements), mutation for genetic algorithms commonly utilizes only random pointwise changes. Furthermore, generalizing beyond pointwise mutations poses a key difficulty as useful genome rearrangements depend on the representation and problem domain. To move beyond the limitations of manually defined pointwise changes, here we propose the use of techniques from masked language models to automatically generate mutations. As a first step, common subsequences within a given population are used to generate a vocabulary. The vocabulary is then used to tokenize each genome. A masked language model is trained on the tokenized data in order to generate possible rearrangements (i.e., mutations). In order to illustrate the proposed strategy, we use string representations of molecules and use a genetic algorithm to optimize for drug-likeness and synthesizability. Our results show that moving beyond random pointwise mutations accelerates genetic algorithm optimization.

BackgroundProtein structure prediction (PSP) has important applications in different fields, such as drug design, disease prediction, and so on. In protein structure prediction, there are two important issues. The first one is the design of the structure model and the second one is the design of the optimization technology. Because of the complexity of the realistic protein structure, the structure model adopted in this paper is a simplified model, which is called off-lattice AB model. After the structure model is assumed, optimization technology is needed for searching the best conformation of a protein sequence based on the assumed structure model. However, PSP is an NP-hard problem even if the simplest model is assumed. Thus, many algorithms have been developed to solve the global optimization problem. In this paper, a hybrid algorithm, which combines genetic algorithm (GA) and tabu search (TS) algorithm, is developed to complete this task.ResultsIn order to develop an efficient optimization algorithm, several improved strategies are developed for the proposed genetic tabu search algorithm. The combined use of these strategies can improve the efficiency of the algorithm. In these strategies, tabu search introduced into the crossover and mutation operators can improve the local search capability, the adoption of variable population size strategy can maintain the diversity of the population, and the ranking selection strategy can improve the possibility of an individual with low energy value entering into next generation. Experiments are performed with Fibonacci sequences and real protein sequences. Experimental results show that the lowest energy obtained by the proposed GATS algorithm is lower than that obtained by previous methods.ConclusionsThe hybrid algorithm has the advantages from both genetic algorithm and tabu search algorithm. It makes use of the advantage of multiple search points in genetic algorithm, and can overcome poor hill-climbing capability in the conventional genetic algorithm by using the flexible memory functions of TS. Compared with some previous algorithms, GATS algorithm has better performance in global optimization and can predict 3D protein structure more effectively.

A comprehensive Genetic Algorithm (GA) model has been developed and applied to derive optimal operational strategies of a multi-purpose reservoir, namely Perunchani Reservoir, in Kodaiyar Basin in Tamil Nadu, India. Most of the water resources problem involves uncertainty, in order to see that the GA model takes care of uncertainty in the input variable, the result of the GA model is compared with the performance of a detailed Stochastic Dynamic Programming (SDP) model. The SDP models are well established and proved that it takes care of uncertainty in-terms of either implicit or explicit approach. In the present study, the objective function of the models is set to minimize the annual sum of squared deviation from desired target release and desired storage volume. In the SDP model the optimal policies are derived by varying the state variables from 3 to 9 representative class intervals, and then the cases are evaluated for their performance using a simulation model for longer length of inflow data, generated using a Thomas–Fiering model. From the performance of the SDP model policies, it is found that the system encountered irrigation deficit, whereas GA model satisfied the demand to a greater extent. The sensitivity analysis of the GA model in selecting optimal population, optimal crossover probability and the optimal number of generations showed the values of 150, 0.76 and 175 respectively. On comparing the performance of SDP model policy with GA model, it is found that GA model has resulted in a lesser irrigation deficit. Thus based on the present case study, it may be concluded that the GA model performs better than the SDP model.

In response to the problems of traditional genetic algorithms in evaluating English compositions, the stability of automatic grading of English compositions has been further enhanced. This article evaluates the teaching effectiveness of automatic grading of English compositions using an optimization fusion algorithm combined with genetic optimization algorithm and CNN model. By analyzing genetic content and optimization algorithms, a corresponding fusion optimization model was obtained, and the automatic evaluation of English compositions was analyzed and predicted through experimental verification. The results indicate that the curves corresponding to different parameters exhibit typical segmentation features through the variation curves of individual numbers under different scale factors. And through quantitative description and analysis of the curve, it can be seen that the change in proportion factor has an absolute advantage in the impact of genetic algorithm on the number of children. As the number of samples increases, the performance of genetic optimization algorithms under the f function shows an upward trend. Research has shown that the writing content index has the greatest impact on English writing, while the corresponding grammar errors have the smallest impact on English writing. Finally, the accuracy of the optimized model was verified by comparing the model curve with experimental data. This study provides theoretical support for the use of genetic optimization algorithms and CNN models in English, and provides ideas for the use of optimization algorithms in other fields.

The scheduling problem is a typical time table problem in educational administration. For such a NP complete problems, when the genetic algorithm solves this problem, it has precociousness phenomenon and quickly converges not to the global optimal solution but to the local optimal solution. Therefore, we use the advantage of local search method and partial matching crossover operation to solve TTP problem with the genetic algorithm combining search algorithm. We do a lot of experiments and evaluate the performance of the improved genetic algorithm. The experiment results show that genetic search algorithm is a more superior algorithm to apply to the TTP problem.

The laws of classical mechanics and quantum theory do not distinguish between the past and the future. But a lot of phenomena tell us that the "arrow of time" points only in one direction. For example, in biology we discover evolutionary processes leading to diversification and increasing complexity; it is a typical feature of such evolutionary processes that "forward" in time is possible but "backward" in time is impossible. At what level can phenomena like these be described? There is another problem in connection with time: within usual quantum theory, space-coordinates and time are not symmetrical to each other, which is in contrast to the basic results of special theory of relativity. In usual quantum theory coordinates are operators and play the role of statistical quantities; time however does not behave statistically and remains a simple parameter when we go from classical to quantum mechanics. For this reason there is no uncertainty relation for the time and energy within usual quantum theory which would agree in its physical content with that for the coordinates and momenta. The purpose of this paper is to show, on the basis of new aspects, that both one-sidedness of time and its uncertainty are charateristics at the fundamental level of quantum theory. We have extended usual quantum theory on the basis of space-time arguments. In this connection a time-operator appeared and there is in general temporal asymmetry at the microscopic level. The picture might be important in the description of phenomena with an "arrow of time", i.e., processes which belong to the physics of becoming. Furthermore, we have derived an uncertainty relation for the energy and time which agrees in its physical content with the uncertainty relation for the momentum and the coordinate; this relation is different from that of usual quantum theory. Within the frame of this physical picture all the concepts (diversity, natural selection, mutation), which are presently used in connection with the evolution of biological systems, are already established at the quantum mechanical level and, therefore, our theoretical picture might also be the foundation for the explanation of the phenomena at the biological (macro-molecular) level.

With the development of science and technology, there has been growing the grid size, the growing demand for electricity and the extent of increasing the electricity market.In this paper, a mathematical morphology and genetic algorithm based dynamic reactive power optimization method for distribution network is proposed.By means of constructing mathematical morphology filter, the binary image composed by allelic genes of chromosome is filtered and the problem that deals with the action times of voltage regulating devices for compensation is turned into filtering problem of discrete binary image.With further automate scheduling, distribution automation and unmanned substation reactive power optimization of distribution line running made an urgent request.Reactive power optimization can improve the voltage passing rate, increasing the reliability of the system, the power quality, and the system runs the security and economy of the perfect combination, very broad application prospects.

Improved random selection strategy, self-adaptive crossover and mutation operators, population co-evolution algorithm are adopted in this paper to set up a new genetic algorithm(GA), guaranteeing the universality of population, the thoroughness of solution and the execution efficiency of algorithm in solving multi-process, process crossover, work-piece flow and other condition based scheduling problem of flexible manufacturing system(FMS). The problem of GA in solving FMS scheduling is summarized in this paper, reasons and countermeasures for GA's pre-maturing are studied, longitudinal and transverse comprehensive state-based correlative self-adaptive crossover and mutation operators are raised according to the analysis on GA to avoid local optimization and pre-maturity, the co-evolutionary GA is designed to solve the scheduling model for FMS, and the algorithm is validated for the scheduling of box-type parts for precise machine tools.

Although fuzzy logic controllers and expert systems have been successfully applied in many complex industrial processes, they experience a deficiency in knowledge acquisition and rely to a great extent on empirical and heuristic knowledge which, in many cases, cannot be objectively elicited. Among the problems to be resolved in fuzzy controller design are the determination of the linguistic state space, definition of the membership functions of each linguistic term and the derivation of the control rules. Some of these problems can be solved by application of machine learning. First, it is desirable to simplify and automate the specification of linguistic rules. Secondly, it is also desirable that modification of control rules is possible in order to cope with previously unknown or changes in process dynamics. Machine learning methods have, in recent years, emerged from the use of learning algorithms modelled on natural and biological systems. These methods attempt to abstract the advanced mechanisms of learning exhibited by such systems, which can consequently be applied to intelligent control. One of these new algorithms is the genetic algorithm which is modelled on the processes of natural evolution. The paper develops the application of genetic algorithm techniques for fuzzy controller design. Genetic algorithms are used to automate and introduce objective criteria in defining fuzzy controller parameters.

ABSTRACTIn order to reduce the venting noise within a confined room, the development of a high performance muffler within certain space constraints is imperative. Therefore, the main purpose of this paper is to develop an optimally shaped one-chamber cross-flow perforated muffler that will dramatically increase the acoustical performance within a limited space.On the basis of plane wave theory, a four-pole system matrix for evaluating acoustical performance is derived by using a decoupled numerical method. Moreover, in order to search for a global optimum, a genetic algorithm (GA) with tournament selection in elitism and uniform crossover/mutation (analogous to a genes' evolutionary process) has been used for the mufflers' optimization. To assure the GAs' correctness, the STLs' maximization of one-chamber cross-flow perforated mufflers with respect to three targeted frequencies (200, 400, and 800Hz) is exemplified. Furthermore, a numerical case for dealing with an octave band noise emitted from an air compressor using one-chamber cross-flow mufflers has been introduced and fully discussed. To achieve a better optimization with the GA, various GA parameter sets were used. Before the GA operation is carried out, the mathematical models of cross-flow perforated mufflers are verified by comparing the numerical result and experimental data with those obtained by Sullivan and Wu respectively.Consequently, the results reveal that the acoustical performance for a cross-flow perforated muffler is excellent. Moreover, they reveal that the optimization approach proposed in this study is easy, economical, and quite effective.

The stochastic and nonlinear characteristics of electric arc furnaces (EAFs) lead to power quality challenges in the power system. In studying EAF behaviors, having optimized characteristics/models, selecting a suitable and optimum model that adapts to the actual characteristics of EAFs, and investigating simulation software’s capability for implementing EAF models are essential. However, the literature shows a research gap in investigating EAF simulations in various software products based on different models. This paper studies several time-domain models, such as piece-wise linear, modified piece-wise linear, hyperbolic, exponential, and exponential-hyperbolic models, for EAF modeling and simulation. The optimal estimation of parameters for the introduced models is necessary to adapt actual EAF characteristics. Thus, one of the studies taken in this paper is optimizing the EAF model’s characteristics. The proposed optimization problem is solved using the genetic algorithm (GA) and particle swarm optimization (PSO). Moreover, the optimized models are simulated in DIgSILENT and EMTP-RV to investigate different EAF models from the viewpoint of accuracy and efficiency. The optimization of different EAF models’ characteristics and comparison of EMTP-RV and DIgSILENT in simulating EAF behavior are the contributions of this paper. The proposed method is validated based on the actual data of a realistic EAF-based steel company in Iran. The obtained results show that the modified piece-wise linear model has the most accuracy in identifying the EAF behavior. The test results based on DIgSILENT and EMTP-RV simulations imply that the EAF could be simulated with high accuracy using modified piece-wise linear and piece-wise linear models. In general, EMTP-RV has expressed more accuracy in simulating different EAF models, and the simulation execution speed of EMTP-RV is around 2.5 times faster than DIgSILENT. In contrast, DIgSILENT is more suitable to facilitate the power system studies of EAF according to its extensive study tools and library.

(2006). Evolution of Biological Systems in Random Media: Limit Theorems and Stability. Journal of the American Statistical Association: Vol. 101, No. 473, pp. 397-397.

The application of genetic algorithms in behavioural ecology, illustrated with a model of anti-predator vigilance

1. Animals foraging for resources are under a variety of selective pressures, and separate optimality models have been developed predicting the optimal reproductive strategies they should adopt. 2. In most cases, the proximate behavioural mechanisms adopted to achieve such optimality goals have been identified. This is the case, for example, for optimal patch time and sex allocation in insect parasitoids. However, behaviours modelled within this framework have mainly been studied separately, even though real animals have to optimize some behaviours simultaneously. 3. For this reason, it would be better if proximate behavioural rules were designed to attain several goals simultaneously. Despite their importance, such multi-objective proximate rules remain to be discovered. 4. Based on experiments on insect parasitoids that simultaneously examine their optimal patch time and sex allocation strategies, it is shown here that animals can adopt multi-objective behavioural mechanisms that appear consistent with the two optimal goals simultaneously. 5. Results of computer simulations demonstrate that these behavioural mechanisms are indeed consistent with optimal reproductive strategies and have thus been most likely selected over the course of the evolutionary time.

Female greater horseshoe bats form maternity colonies each summer in order to give birth and raise young. During the mating period, females visit males occupying territorial sites, copulation takes place and sperm are stored until ovulation occurs, normally in April. Using microsatellite markers and a likelihood method of parentage analysis, we studied breeding behaviour and male reproductive success over a five-year period in a population of bats in south-west Britain. Paternity was assigned with 80% confidence to 44% of young born in five successive cohorts. While a small annual skew in male reproductive success was detected, the variance increased over five years due to the repeated success of a few individuals. Mating was polygynous, although some females gave birth to offspring sired by the same male in separate years. Such repeated partnerships probably result from fidelity for either mating sites or individuals or from sperm competition. Females mated with males born both within and outside their own natal colony; however, relatedness between parents was no less than the average recorded for male female pairs. Gene flow between colonies is likely to be primarily mediated by both female and male dispersal during the mating period rather than more permanent movements.